OXFORD CLASSICAL MONOGRAPHS

Published under the supervision of a Committee of the Faculty

of Classics in the University of Oxford
 The aim of the Oxford Classical Monograph series (which replaces the
Oxford Classical and Philosophical Monographs) is to publish books based
on the best theses on Greek and Latin literature, ancient history, and
ancient philosophy examined by the Faculty Board of Classics.
Aristotle’s Categories
in the Early Roman
Empire

M IC HA E L J . G RI F F IN

1
 3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
# Michael J. Griffin 2015
The moral rights of the author have been asserted
First Edition published in 2015
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2014951230
ISBN 978–0–19–872473–5
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
 Parentibus carissimis
‹’ r  K ŒØ KçÅ Æ ŒFØ
Proclus ap. Olymp. in Phaed. 1.5,16
 Preface

This study began life as a dissertation submitted to the University of

Oxford in 2009. That dissertation aimed, rather ambitiously, to settle
a problem that Hans Gottschalk has neatly articulated as follows: ‘[i]t
would be interesting to know why the Categories came to exercise so
much fascination [in the first century bce], but there is no evidence’
(1987: 1103). Building on the later Neoplatonic sources, especially
Simplicius (c.490–c.560 ce), I hoped to provide evidence for a solu-
tion of that puzzle. Along the way, I reconstructed the long and
complex stratigraphy of commentary on the Categories before Por-
phyry, following (or at least muddling along) in the footsteps of
scholars such as Paul Moraux (1973, 1984).
The present book differs from that dissertation in several respects.
Most importantly, it has benefited from several years of critical
feedback. It has also, for better or worse, been pared down to focus
on a single story. Andronicus of Rhodes, I suggest, rescued the
Categories from obscurity because he believed it was a good begin-
ner’s introduction to Aristotle’s theory of demonstration (I Ø Ø).
But contemporary and later philosophers’ confusion regarding its
subject-matter—language, concepts, or reality?—helped to draw fur-
ther attention to the treatise, fostering the development of subsequent
ancient semantics. Those chapters that did not especially contribute to
this story have been deleted, while the new chapter 2 (Andronicus of
Rhodes) has been heavily revised and expanded. Chapter 3 (Eudorus and
Pseudo-Archytas), chapter 4 (Lucius and Nicostratus), and chapter 5
(Athenodorus and Cornutus) have been moderately revised, while
chapter 6 (Boethus of Sidon) has been reduced in scope, since some of
my conclusions in 2009 have been corrected, or improved, by a series
of recent, outstanding articles and edited collections on this period (see for
example Rashed 2013, Chiaradonna 2013). At the time of writing, I was
able to make only limited use of the new commentary on the Categor-
ies found in the Archimedes Palimpsest (Chiaradonna, Rashed, and
Sedley 2013), but I have been able to add references to several sup-
porting passages from that text.
I have incurred many debts of gratitude in the development of this
project. I am especially indebted to Tobias Reinhardt, under whose
viii Preface
kind and patient supervision the dissertation gradually wound its way
into being, and to Richard Sorabji and Peter Adamson, who carefully
examined the entire manuscript at the dissertation stage, and have
continued to discuss revisions with me in detail. I am also grateful to
Professor Sorabji for many years of conversations about the philoso-
phy of the commentators, which have greatly expanded my know-
ledge and appreciation of the period. The dissertation project was
originally suggested by Michael Frede, my faculty adviser for the
M.Phil. at Oxford from 2004 to 2006, without whom it would never
have begun. Ben Morison, who kindly agreed to serve as my mentor
for the Press, has been truly generous with his time and insightful
commentary. I am also indebted to several careful readers of multiple
drafts of the MS or individual chapters, including Jonathan Barnes,
Sylvia Berryman, George Boys-Stones (who also kindly shared with
me helpful unpublished work), Michael Chase, Riccardo Chiara-
donna, Myrto Hatzimichali, George Karamanolis, Stephen Menn,
Marwan Rashed, David Sedley, Robert Sharples, Richard Sorabji, and
Thomasz Tiuryn, as well as participants in conferences and lectures
held at the University of Toronto, University College London, and
the University of Trier. I am also grateful to Annie Hewitt for her
painstaking and careful editorial work on the manuscript, which has
significantly improved the final product and saved me from not a few
embarrassing errors. None of these readers, of course, can be faulted
for the remaining mishaps and errors, which are solely my own.
My graduate work at Oxford was generously supported by the
Commonwealth Scholarship Commission of the United Kingdom
and the Social Sciences and Humanities Research Council of Canada.
The development of this book was also sponsored by an Insight Grant
from the Social Sciences and Research Council of Canada. I am very
grateful to both bodies for their kind support from the beginning to
the end of this process. Last but by no means least, I would like to
thank my parents, Bruce and Jennifer, who have made everything
possible, and sustained this effort from archē to telos.
Michael J. Griffin
The University of British Columbia
December 2013
 Contents

Abbreviations xii

1. Introduction 1
I. This Study: The Categories in the Early Roman Empire 7
Road Map 8
Goals, Methods, and Materials 14
II. The Structure of the Categories 16

Part A. Rediscovery and Endorsement:

Andronicus and Eudorus
2. Andronicus of Rhodes 21
I. Life and History 23
II. Works and Editorial Activity: Beginning with the
Categories 29
III. Andronicus on the Categories 32
IV. Conclusions: Andronicus’ relationship to the ‘ancient
exegetes’ of Categories 72
3. ‘Pythagorean’ Support: Eudorus and Pseudo-Archytas 78
I. Eudorus of Alexandria (fl. c.50 bce?) 78
Life and History 78
Works 79
Philosophy 80
Treatment of the Categories 82
Relationship to Andronicus 95
II. Pseudo-Archytas (fl. c.50 bce?) 97

Part B. Early Criticisms: Platonists and Stoics

4. Lucius and Nicostratus 103
I. Using the Neoplatonist Sources 106
II. Life and History 108
Lucius 108
Nicostratus 111
III. Distinct Portrayal as Critics 112
x Contents
IV. Questions and Responses 113
Respondents to the Nicostratans 115
Respondents to the Lucians 118
Summary of Treatment by Respondents 126
5. Stoic Critique: Athenodorus and Cornutus 129
I. Life and History 135
Athenodorus 135
Cornutus 139
II. Sources for Two ØÆ æØ in the Categories 145
Athenodorus and Cornutus on the ‘Division’ of
the Ten Categories 147
Athenodorus and Cornutus on the ‘Division’
of Chapters 165
III. Conclusions 171

Part C. Peripatetic Synthesis and Response

6. Boethus of Sidon 177
I. Life 179
Relationship to ‘Lucius and Nicostratus’ 180
Relationship to Andronicus 181
II. Works 183
Evidence for Line-by-Line Study of the Categories 185
Responses to I æ ÆØ about the Categories 186
III. On the Π of the Categories 187
The Categories is about Simple, Significant Terms
(and Concepts) 187
The Categories is about Sensible Beings
(the Referents of Simple, Significant Terms) 189
IV. The Structure of Boethus’ Work: I æ ÆØ and ºØ 193
V. Conclusions 198
7. The Second Century: A Brief Overview 200
I. Second-Century Platonists 200
II. Peripatetics 201
Alexander of Aphrodisias, His Teachers and Contemporaries 202
8. Conclusions 205
 Contents xi
Appendices 209
1: Persons and Sources 211
I. Main Persons Treated 211
II. Main Sources 216
2: Andronicus’ Publication and Works 219
I. Publication and Catalogue of the Aristotelian
Corpus 219
Sources and Discussion 219
Assessment of Andronicus’ Role 223
II. Other Works of Interest 229
Andronicus’ Book ‘On Division’ 229
Andronicus’ Paraphrase of the Categories 231
Andronican Spuria 232
Letters of Aristotle 232
3: An Outline of Aristotle’s Categories 233

Bibliography 243
General Index 273
Index of Passages 278
 Abbreviations

ANRW Aufstieg und Niedergang der römischen Welt

CAG Commentaria in Aristotelem Graeca
Cat. Categories
DL Diogenes Laertius
DPhA Goulet, R. Dictionnaire des philosophes antiques. Volumes as
cited in bibliography.
in Cat. On the Categories
LS Long, A. A., and Sedley, D. N. (1987). The Hellenistic
Philosophers. Vols I–II. Cambridge: Cambridge University Press.
LSJ Liddell, H., Scott, R., and Jones, H. (1940). A Greek–English
Lexicon (9th edn). Oxford: Oxford University Press.
Moraux I Moraux, P. (1973). Der Aristotelismus bei den Griechen von
Andronikos bis Alexander von Aphrodisias—Erster Band: Die
Renaissance des Aristotelismus im I. Jh. v. Chr. (Vol. I). Berlin/
New York: Walter de Gruyter.
Moraux II Moraux, P. (1984). Der Aristotelismus bei den Griechen von
Andronikos bis Alexander von Aphrodisias—Zweiter Band: Der
Aristotelismus im I. und II. Jh. n.Chr (Vol. II). Berlin/New York:
Walter de Gruyter.
OSAP Oxford Studies in Ancient Philosophy
Porph. Porphyry (normally in Cat.)
RE Wissowa, Real-Enzyclopädie der klassischen Altertums-
wissenschaft (Stuttgart and Munich, 1894–1980).
Simplic. Simplicius (normally in Cat.)
SVF H. von Arnim, Stoicorum Veterum Fragmenta, i–iii (Leipzig,
1903–5; Vol. IV, index, compiled by M. Adler, 1924).
—Well, then, what is the reason that impelled the philosophers
of old to engage in disputes of every kind with each other on the
subject of this Aristotelian writing which we call the Categories?
For as far as I can see, neither have more numerous controver-
sies occurred about any other topic, nor have greater contests
been stirred up, not only by Stoics and Platonists trying to
undermine these Aristotelian Categories, but even among the
Peripatetics with each other . . .

—It is because . . . the subject (Œ ) of this book concerns the

primary and simple expressions (º Ø) <and the realities> they
signify. So since º ª is useful to all branches of philosophy,
and the first principles of this are simple expressions and their
objects of reference, it is natural that much controversy has
arisen . . .
Dexippus, On the Categories 5,16–22
(tr. after Dillon 1990), fourth century ce

Es war ein eines scharfsinnigen Mannes würdiger Anschlag des

Aristoteles, diese Grundbegriffe aufzusuchen. Da er aber kein
Prinzipium hatte, so raffte er sie auf, wie sie ihm aufstießen, und
trieb deren zuerst zehn auf, die er Kategorien (Prädikamente)
nannte. In der Folge glaubte er noch ihrer fünfe aufgefunden zu
haben, die er unter dem Namen der Postprädikamente hinzu-
fügte. Allein seine Tafel blieb noch immer mangelhaft.

It was an enterprise worthy of an acute thinker like Aristotle to

try to discover these fundamental concepts; but as he had no
guiding principle he merely picked them up as they occurred to
him, and at first gathered up ten of them, which he called
categories or predicaments. Afterwards he thought he had dis-
covered five more of them, which he added under the name of
post-predicaments. But his table remained imperfect for all
that . . .
Kant, Critique of Pure Reason Transcendental
Doctrine of Elements, A81/B107
(tr. N. Kemp Smith)
There is a theory called the theory of categories which in a more
or less developed form, with minor or major modifications,
made its appearance first in a large number of Aristotelian
writings and then, under the influence of these writings, came
to be a standard part of traditional logic, a place it maintained
with more or less success into the early part of this century,
when it met the same fate as certain other parts of traditional
logic.
Frede 1987: 29
 1

Introduction

The short, compressed philosophical essay that comes down to us

under the name Categories (˚ÆŪæÆØ) has exercised a special kind
of fascination throughout the Western intellectual tradition. For two
millennia of recorded study, the Categories was embedded in the
fabric of philosophical education as an introduction to logic.1 It
played a crucial role in the Greek, Latin, and Arabic philosophical
traditions, and subsequently received a thoughtful critique from the
seminal philosophers of early modern Europe, among them Des-
cartes, Spinoza, Leibniz, and Kant.2 The Categories retired from its
place as the gatekeeper of philosophy and higher education alongside
the rapid transformation of modern European logic in the nineteenth
century, but even its retirement years have been busy and fruitful:
modern students of Aristotle have come second to none of their
predecessors in working out novel and careful interpretations of the
place of the Categories in Aristotle’s philosophical system and intel-
lectual development,3 and it continues to inspire new ideas, especially
in metaphysics.4
The Categories’ puzzling organization contributes to its difficulty
and allure. As we have it today, the treatise is divided into three
traditional sections. (1) The first section begins with a description of
several ways in which multiple referents can share a name in ordinary
language (ch. 1). The author then partitions meaningful speech into
complete sentences and referring terms (ch. 2). Finally, he proceeds to

1
Frede 1987: 29.
2
See Gillespie 1925: 79 = Barnes, Schofield, and Sorabji 1979: 1.
3
See for example Mann 2000, Wedin 2000, Bodéüs 2001: xi–clxxxviii, Barnes
2005a, and the contributions of Bruun and Corti 2005.
4
See for example Lowe 2006.
2 Introduction
an influential division of being, which privileges individuals (items that
are not the kinds, attributes, or modes of anything else) as exhibits of
primary reality or substance (PÆ) (ch. 3). (2) The second section,
which is often treated as the heart of the treatise, lists ten items, perhaps
kinds of predicates or predications (chs. 5–9). This list is comparable
to the list that occurs at Topics 1.6, although there are differences.
Here the list comprises reality (for example, man, horse), quantity
(four-foot, five-foot), quality (white), relative (double, half), where
(in the marketplace), when (yesterday, last year), in-position (lying
down), having (shod, armed), acting (cutting, burning), and being-
acted-on (being cut, being burned). Finally, (3) the third traditional
section, which has occasionally (since antiquity) been viewed as an
intrusion stitched into the treatise by a well-meaning editor, examines
concepts like opposition, priority, and change. All three sections have
attracted debate.5 (See below, ‘The Structure of the Categories’, and
Appendix 3 for a more detailed account of the content of the book).
The Categories offers a series of interesting, separated reflections,
some of which appear to belong to ontology, others to logic or to the
grammar of a natural language or to semantics. This topical diversity
has sometimes been interpreted as implying a systemic confusion of
use and mention on Aristotle’s part;6 I argue in this book that it also
led ancient philosophers to sharpen their own conceptual distinctions
between these different sub-disciplines of philosophy. But who would
compose such a treatise in the first place, and to what end? The
Categories has certainly not come down to us in the form that
Aristotle intended—if indeed Aristotle was the author of some of

5
For example, readers have continually puzzled over the compatibility of the
Categories’ account of primary PÆ with the hylomorphism of the central books
of Aristotle’s Metaphysics (see Simplicius, in Cat. 78,4–20). There has rarely been a
clear consensus about the origins of the central list of ten items, or about their nature
(words, concepts, or things?) Even the Neoplatonist commentators, masters of con-
sensus, were hard pressed to deliver a compromise that really fit the text.
6
Bäck (2000) remarks that the philosopher appears to ‘switch back and forth from
speaking about words to speaking about real things’ and does so ‘quickly and blithely’
with confusing rapidity (133–4). Two recent approaches to this ambiguity are evident
in Wedin (2000) and De Rijk (2002:I). As a policy, Wedin will ‘follow Aristotle’s
practice of disregarding use-mention boundaries in formulating a number of his
theses’ (12 n. 6). By contrast, De Rijk, who expresses this equivocation as ‘the absence
of a clear-cut borderline between a linguistic expression [ . . . ] taken as a linguistic
tool, and its significate’, proposes that Aristotle exploits it intentionally and system-
atically, and christens this ‘the rule of indiscriminate reference’ (63–4)—a necessary
law and ‘main rule’ for his exegetical approach to Aristotelian semantics.
 Introduction 3
the text, which has sometimes been doubted.7 It has almost nothing
explicit to tell us about its own goals as a philosophical treatise or
lecture; nothing about its relationship to other treatises by any author;
and it employs numerous locutions and arguments that do not
occur elsewhere in Aristotle. Some time before the first century bce,
its editors (or its author) had complemented the text with numerous
interpolations in order to construct a coherent treatise. Even the
name Categories or Predications (˚ÆŪæÆØ) was only one of
many titles circulating in antiquity, each of them reflecting a different
vision of the aim or subject-matter of the treatise.
If we do allow, with the consensus of most contemporary scholars,8
that at least the first nine chapter-headings of our Categories derive
from an authentic treatise by Aristotle, then we are left with a number
of difficult problems: for example, what kind of text it is, how
Aristotle meant it to be read, whether it is internally consistent,
when in his complex philosophical development it might have been
composed, how it relates to the other surviving treatises by Aristotle,
and so forth.
The present study does not set out to grapple directly with these
problems, which attract new analysis on a regular basis and may well
continue to do so for another two millennia, f ŁE.9 Rather, this
study explores an essential chapter in the history of how these
problems were approached in antiquity. In particular, it seeks to
recount how the readers of Aristotle during the age of the first

7
In the nineteenth century by Spengel 1845: 41; Prantl 1846: 646 and 1855: 90 n. 5;
Rose 1854: 232; Gercke 1891: 424–41, esp. 437; and more powerfully in the twentieth
century by Dupréel 1909 and Mansion 1949: 1097–1100.
8
For recent discussion of the authenticity of the Categories, see for example
Bodéüs 2001: xi–clxxxviii, Frede 1987: 11–48, Husik 1952: 97–203, and De Rijk
1951. Its authenticity was not questioned in antiquity, with the exception of the
postpraedicamenta. It is broadly accepted as Aristotelian today, and discussed as
Aristotle’s work by most critics, including for example Ackrill 1963, Barrington
Jones 1972, Dancy 1975, Irwin 1981, Furth 1988, Rist 1989, Lewis 1991, Wedin
2000, Mann 2000, and Shields 2003.
9
The introduction to the edition of Bodéüs (2001: xi–clxxxviii) and the short
essays by Frede (1987: 11–48) provide an excellent overview of issues relating to the
function, title, authenticity, and composition of the treatise. The arguments that we
should understand the Categories primarily as a dialectical handbook related to the
Topics, and not as a contribution to first philosophy, seem strong (see Menn 1995);
but it is also worth stressing its profound contribution to the shape and spirit of
Aristotle’s metaphysics (Mann 2000), whether or not it contains a metaphysical
theory similar to or competitive with that of Metaphysics ZHΘ (see for example
Wedin 2000).
4 Introduction
philosophical commentaries—coincident with the rise and teetering
of the Roman Empire—thought that these problems might have been
solved, and what kinds of solutions they sought to offer, sometimes
in cooperation and sometimes in competition. The interest of their
solutions is intrinsically clear: for it was through their early accounts
of the Categories, both polemical and exegetical, that this koan of a
book initially came to exercise so much fascination in the ancient
philosophical schools, and ultimately made its way to the forefront
of the late antique curricula as a foundation of philosophical study,
with substantial implications for the subsequent course of European
philosophy.
Several studies have demonstrated the value of the history of
commentary on the Categories. Over the centuries, changing treat-
ments of the Categories tracked changing conceptions of logic and the
goal of philosophy. Several scholars have pointed to the first century
bce as the crucible of its fascination. For instance, Robert Sharples
(2008: 274) has remarked that
[t]he attention given to Aristotle’s Categories in antiquity had major
consequences for the future direction of philosophy. The prominence in
subsequent discussion of the problem of universals, and more generally
of questions concerning the relation between being, knowledge and
language, is due in large part to the Categories coming in antiquity
to occupy the place it did at the start of the philosophical curriculum.
This has also affected approaches to Aristotle himself.
Marwan Rashed’s innovative treatment of Alexander of Aphrodisias,
Essentialisme (2007), has shown that Alexander (in the late second
and early third century ce) interpreted Aristotle’s De Anima and
Metaphysics through the lens of the Organon, and especially the
Categories.10 The Neoplatonist Porphyry explicitly defended the
value of the Categories as an introduction to philosophy, and through
his agency, centuries later, it had secured a place at the gateway of
the Neoplatonic curriculum for Aristotelian studies. Subsequent
scholarship on the Categories in the Latin West demonstrates its
far-reaching influence on thinkers ranging from Boethius through
Eriugena (c.800–c.877), Abelard (1078–1142), Aquinas (1225/6–74),

10
Sharples (2008) also points to the work of Arthur Madigan (1994), who suggests
that Alexander read the Metaphysics ‘in light of the Categories rather than vice
versa’ (90).
 Introduction 5
Roger Bacon (1214–94), Duns Scotus (c.1265/6–1308), William
Ockham (c.1285–1347), and into the fifteenth century with Latin
commentators such as Paul of Venice: for a recent overview, see
Newton (2008). In the Arabic tradition, as Peter Adamson notes,
the seminal philosopher and translator al-Kindī (c.800–70) took a
special interest in the Isagoge and the Categories,11 and al-Fārābi
(c.872–c.950) commented on the Organon in full,12 although Avi-
cenna (c.980–1037) would later offer the definitive replacement for
much of the traditional Organon. Nevertheless, even early modern
and current trends in Arabic and Persian philosophy, following
Avicenna, demonstrate as strong an interest in the Organon as may
be found in contemporary Anglo-Saxon philosophy.13
But the chapter of the story that precedes Alexander—that is, the
reception of the Categories during the period spanning the gap from
the alleged ‘rediscovery’ of Aristotle’s esoteric works in the first
century bce to the first extant commentaries on the Categories in
the third century ce—has not yet been treated in a comprehensive
study.14 It was during this period that the Categories came to acquire
its location at the beginning of the Aristotelian curriculum, which
it held by the late second century. And it was also during this
period that many of the familiar later difficulties regarding the Cat-
egories were framed, such as its relationship to the Metaphysics of
Aristotle and the core problem of defining its subject-matter. Of more
general historical interest, perhaps, this period saw a crucial shift
of philosophical activity from a more oral, geographically centralized
practice to a textual, geographically diverse enterprise that relied
increasingly on commentary (see Frede 1999, Boys-Stones 2001,
Hatzimichali 2013). The change is effectively summarized by Sedley
(2012: 1–2):

11
Adamson (2007b), 15, 26–8.
12
McGinnis and Reisman (2007), 65.
13
A recent discussion as to whether the categories properly belong to ontology or
logic (noted in Ziai 2005) demonstrates that the aporiai of antiquity remain vital
across cultures, languages, and centuries.
14
The brief survey in Bodéüs (2001) and Gottschalk (1987), together with the
longer treatment in Moraux (1973, 1984) are all invaluable in introducing the period
and the relevance of the Categories in it; for the historical background, see now
Schofield (2013), and for a selection of primary texts in translation, Sharples (2010).
6 Introduction
The first century bce is the period in which philosophy loosened
its historic moorings in the great philosophical schools of Athens and
entered the Roman world, often attaching itself to such cultural
centres as Alexandria and Rome. It is no exaggeration to say that
the character of philosophy as an intellectual activity was permanently
changed by this transition. Cut adrift from the historic institutions
which had linked them to their revered founders, the major philoso-
phies shifted their efforts increasingly onto the study of their foun-
dational texts. The Roman imperial age was thus an age in which
philosophy centred on the newly burgeoning industry of textual
commentary, above all commentary on the writings of Plato and
Aristotle.
The philosophers with whom the present study is primarily con-
cerned lived between about this time and the middle of the third
century of our era.15 Since, with few exceptions, the lives of the
principal authors are not securely dated, the bounding dates of this
study are rather arbitrary. To keep things simple, in the title I have
referred to the ‘early Roman empire’: our protagonists were active
mainly during and after the beginning of Augustus’ principate in
27 bce, and our story ends after the reign of Marcus Aurelius.
A little more precisely, I attempt to collect and examine the first
preserved fragments of views related to the Categories after Theo-
phrastus, which appear to have been formulated in the first century
bce during a tremendous upsurge of interest in Aristotelian studies
and the original texts of Aristotle’s school-treatises, a transformation
which is frequently associated with the name of Andronicus of
Rhodes. I conclude before the third century, excluding Alexander of
Aphrodisias and Galen as primary subjects, but making use of their
work and later work as sources. My sources include related subject
matter by Alexander, responses to the Categories itself by Plotinus
and Porphyry, followed by Iamblichus and Dexippus in the third and
fourth century, Ammonius and Boethius in the fifth and sixth, and
Simplicius and Olympiodorus in the sixth.

15
Griffin and Barnes (1989), Barnes and Griffin (1997), Long and Sedley (1987),
and Sorabji and Sharples (2007) provide overviews of the literature on the intellectual
life of the early Roman empire and the second sophistic. Authoritative studies on the
Peripatetic school in our period are discussed below.
 Introduction 7
I. THIS STUDY: THE CATEGORIES IN
THE EARLY ROMAN EMPIRE

Hans Gottschalk, in a valuable survey, wrote that ‘[i]t would be

interesting to know why the Categories came to exercise so much
fascination [in the first century bce], but there is no evidence’ (1987:
1103). This study will attempt to reconstruct a tentative narrative of
early interest in the treatise (roughly from the first century bce to the
early third century ce), drawing upon the fragmentary evidence of
later Neoplatonist sources. Part A (chs 2–3) considers evidence for an
initial surge of interest in the Categories, which I credit primarily
to Andronicus of Rhodes in the middle of the first century bce.
Part B (chs 4–5) traces critiques of the treatise by Platonist and
Stoic readers. Part C (chs 6–7) surveys responses to those criticisms
by Boethus of Sidon, and subsequent developments. I will credit the
turning-point in the career of the treatise Categories to Andronicus
(ch. 2) and his emphasis on its value as a pre-technical primer for the
Aristotelian science of demonstration (I Ø Ø).
But I also stress in chs 2–3 the evidence for earlier analysis and
criticism of the doctrine of the ten categories in the Old Academy,
perhaps as a subdivision of the Platonic distinction of Absolute from
Relative:16 I argue in ch. 3 that Eudorus of Alexandria drew upon that
Academic tradition during the first century bce and promoted a ten-
categorial scheme compatible with it, and that either Eudorus or
Andronicus may be independently responsible for resurrecting a
‘categorialism’ attributed to both the Old Academy and Aristotle.
Andronicus in turn published a new catalogue of Aristotle’s technical
works that foregrounded the treatise Categories, which may in turn
have earned it a new readership.
The history of the reception of the treatise begins before Androni-
cus and Eudorus. It may have been known in the Hellenistic period
under the title Before the Topics,17 either as a rhetorical or dialectical
handbook.18 In this time, it might have been criticized as a poor

16
Sophist 255D; cf. Xenocrates fr. 12 Heinze/95 Isnardi Parente.
17
—æe H  ø, for which see Simplicius in Cat. 379,9–11, Andronicus ap.
Boethium in Cat. 263B, Porphyry in Cat. 56,22–31. —æe H  ØŒH may have been
preferred by Adrastus (Simplic. 15,30, 16,14).
18
For the later Peripatetic division within logic between persuasive dialectic and
rhetoric (on the one hand) and demonstrative analytic (on the other), see for example
DL 5.28; for a similar Stoic subdivision, see DL 7.43, and for a Middle Platonist
8 This Study
organization of linguistic items (like nouns, verbs, and conjunctions)
by Athenodorus, a Stoic who interpreted the treatise as an unsophis-
ticated contribution to rhetorical or linguistic theory.19 In the first
half of the first century bce, scholarly interest developed in Aristotle’s
‘esoteric’ corpus (that is, the school-treatises which we possess today,
in contrast to the popular dialogues that we have now largely lost),20
represented in the activities of bibliophilic men like Apellicon and
Tyrannio (Strabo, Geog. 13.1.54). Several philosophers, prominently
including Andronicus and Ariston, embarked on a serious study of
the texts, resulting in Andronicus’ much-debated publication, the
putative ‘Roman edition’ of Aristotle and his influential catalogue
( ƌ) of the books of Aristotle.

Road Map
The story of the treatise Categories, as I argue in ch. 2, properly begins
with Andronicus, who played a crucial role in bringing the study of
the treatise into vogue. His new catalogue, which soon became a
standard reference (Plutarch, Sulla 26.1–2, Porphyry VP 24.7), fore-
grounded the text that had previously been obscure, and known
under the title Before the Topics. Andronicus retitled this treatise
˚ÆŪæÆØ, ‘Predications’, and brought it to the front of his
reading list. He did so, I suggest, in order to draw attention to what
he regarded as its pedagogically and epistemologically valuable con-
tent and methods. Andronicus found the treatise useful, not merely as

version, see Alcinous, Handbook 3.2. The Categories might have been read as offering
a source for a handbook division of kinds of questions, similar to the ten elementa
represented in Quintilian Inst. 3.23–34, or already (as I argue Andronicus would later
read it) as a useful heuristic guide for finding the genus of a given term (cf. Menn
1995), supporting defences against fallacies based on category errors like those
represented in Aristotle, Soph. El. 22 and LS 37H (a coat is manly, but to be manly
is brave, so the coat is brave; or again, a body is white, but white is a colour, so a body
is a colour; cf. Simplicius in Cat. 24,6–25,10, Bobzien 2005, Atherton 1993: 180–3).
19
See for example Porphyry, in Cat. 18,26 and 59,5–14, Simplicius in Cat. 62,24–8,
and ch. 5 below. Both Athenodorus and Cornutus seem to have critiqued the text’s
division of various lexical categories, as if Cat. set out to be a textbook of ‘logic’ in the
Stoic sense at the first (rhetorical or grammatical) level, rather than the second
(dialectical) level (cf. DL 7.43, and LS }31). I also attempt to offer some arguments
in favour of this interpretation in Griffin 2013b; see ch. 5 below.
20
For more on the distinction between the esoteric and exoteric treatises by
Aristotle, see ch. 2 below.
 Introduction 9
a handbook for dialectical engagement (see for example Simplicius in
Cat. 379,9–11),21 but as a preliminary to genuine scientific demon-
stration (I Ø Ø; cf. Philoponus, in Cat. 5,15–20). On Andronicus’
view, as I will argue, the Categories introduces us to the genera of
predication in a ‘pre-philosophical’ manner appropriate to beginners,
by deploying sketch accounts or delineations ( ªæÆçÆ) to tease
out or ‘articulate’ ( ØÆæŁæF) our untutored preconceptions
( æºłØ) of the ten genera,22 drawing on the resources available
in ordinary language before we can carve out a more exact and
scientific vocabulary. Thus Andronicus valued the Categories for its
ability to introduce us to predication (ŒÆŪæÆ) in a pre-philosoph-
ical and non-technical manner.23 More precisely, the Categories
helps us to distinguish between essential and accidental kinds of

21
See also Boethius in Cat. 263B. Andronicus jettisons the final five chapters or
‘post-praedicamenta’ for this reason (Cat. }}10–15 retail subjects like contrariety,
priority, and the different senses of ‘having’, which Andronicus judges irrelevant to
the primary function of the text, although useful for approaching the dialectical
methods of the Topics).
22
I ascribe this argument to Andronicus as well as Boethus of Sidon; thus I suggest
that Andronicus relies on a basically Stoic and Epicurean account of the nature of
preconceptions (for the role of such pre-definitional accounts in Hellenistic thought,
see for example Brittain 2005). For my attempt to trace this argument to the first
century bce and to Boethus of Sidon, see in particular Porphyry in Cat. 59,17–33,
Herminus’ second-century ce ‘compressed’ ( Øa æÆåø) articulation of the view that
the Categories caters to the young (Ø, 59,21) and teases out the sorts of precon-
ception ( æºłø, 59,27) that are a pre-philosophical prerequisite for a more
rigorous study of the genera of predication, beginning from the sunētheia or ordinary
language available to hoi polloi. Porphyry has just previously (59,17–18) suggested
that Herminus is compressing the view adopted earlier by Boethus of Sidon. See also
Simplicius in Cat. 159,10–15, where Boethus also argues that it was not possible to
give proper, formal definitions of the primary genera at the early stage represented by
Cat., but that Aristotle provides ªæÆçÆ that can actuate or ‘stir up’ (IÆŒØE) our
concept (ØÆ) of a category like the Relative. For the further attribution to Andro-
nicus behind Boethus, my argument also draws on passages like Simplicius in
Cat. 21,15–22,1, where Andronicus is associated with the necessity of a æ ºÅłØ of
homonyms which the Categories articulates, and Porphyry’s response (379,12–20) to
Andronicus’ disconnection of the post-praedicamenta at 379,8–11, where the argu-
ment against Andronicus assumes that he, like Porphyry himself, is committed to
regarding the treatise, especially in the ‘onymies’, as teasing out (‘articulating’,
ØæŁæøØ) our æºłØ; and 154,3–6, where Porphyry follows Andronicus in
‘articulating’ our ØÆ of the unit or one in treating the category of quantity.
23
For instance, if I would like to know whether Socrates is an individual substance,
Cat. 2a11–14 provides rules to test whether I apply the name ‘Socrates’ as if it labelled
an entity said-of or in another (cf. Simplicius in Cat. 75,23–76,1).
10 This Study
predication24 (without having trained in logic in advance)25 in order
that we can begin to study division ( ØÆæØ) and eventually con-
struct proper definitions based on essential features of the object
rather than its accidental attributes.26 Andronicus interprets the
Categories’ fourfold division (1a20–b9) and its sketch of the attributes
that characterize substances (PÆØ, 2a12–4b19) in contrast to non-
substantial predicates (4b20–11a39) as a useful framework for apply-
ing Aristotelian essentialism in the practice of dialectic. That is, he
employs the structure of predication in the Greek language as a rough
guide to the structure of reality.
What does that mean, exactly? A student who grasps the Categor-
ies’ lessons will be equipped to distinguish the features that mark
a thing out for what it is, and so properly belong in its definition,
from those incidental features that merely describe how it is that
way.27 (Indeed, Andronicus will criticize ‘earlier’ Peripatetics for
failing to recognize the importance of this distinction in dialectic
and definition28—much as later Aristotelians would take the Stoics
to task for failing to distinguish between an essential definition and a
proprium,29 perhaps implying that Stoic dialectical practices were a

24
So the attribution to Andronicus by Boethius (evidently following Porphyry) of
praise for the usefulness (utilitas) of division (De Div. 4,3–5 Magee); Andronicus also
seems to have cultivated the recognition of the distinction between secundum se and
secundum accidens predication in division (De Div. 48,26–50,5, if Andronicus is to be
identified, as Magee argues ad loc., with posterior . . . Peripatetic secta at 48,26). And,
as I argue below, this seems like the right context in which to understand Andronicus’
subsumption of the ten categories under the ‘absolute’ and the ‘relative’ (Simplic. in
Cat. 63,22–8). On Andronicus in the De Divisione, see now Magee 1998: xxiv–lvii; on
Andronicus’ use of the absolute-relative division from Sophist 255D, see Reinhardt
2007. The later tradition treats the division of substance and accident as the ‘primary
division’ in the Categories; see for example Alexander in Metaph. 242,15–243,3
fi
(B æfiÅ ØÆØæØ H K ˚ÆŪæÆØ).
25
Instead offering heuristic tests (presupposed by other works in the Organon) for
determining the genus to which a term belongs. See Bodéüs 2001: xxxix and Menn
1995: 319–20, to which I refer often below.
26
Compare Aristotle An. Post. 2.13, Boethius De Div. 18,4–19 Magee (and follow-
ing) for the principle, and 34,16–36,26 for an extended sample (although there is no
good reason to suppose that this sample is Andronican).
27
For a similar essentialist reading of the Categories, see Loux 1991: ch. 1 (33), and
Sirkel forthcoming. For the Categories as a key text elaborating the distinction
between ‘things’ and their features, see Mann 2000.
28
In Boethius De Div. 891–2, 48,26–50,5 Magee; see ch. 2 below for the attribution
to Andronicus.
29
Alexander in Top. 42,27–43,2, LS 32E. This is not to say that the Stoics really did
fail to make such a distinction, only that they were perceived as vulnerable on this
point by some Peripatetics after Andronicus.
 Introduction 11
target of Andronicus’ critique and that his new emphasis on the
Categories would provide an essentialist framework for demonstra-
tion). In addition, Andronicus treats the clear distinction between per
se and per accidens predication as a prophylactic against various
fallacies that trade on equivocation,30 some of which had been used
earlier in the first century to threaten the value of dialectic (compare
Cicero, Academica 2.91–2). Thus the Categories is rightly re-titled
ŒÆŪæÆØ, on Andronicus’ view, because it helps us to ‘predicate’
successfully—that is ‘to apply a simple significant expression to what
it signifies’31—as a key building-block in the construction of a scien-
tific definition. The curriculum builds from the correct predication of
such ‘uncombined’ or simple terms through the study of sentences
and syllogisms to the art of demonstration taught in the Analytics,32
and this is reflected in Andronicus’ proposed version of the opening
lines in his paraphrase, which makes the treatise about simple or
‘uncombined’ expressions, the proper treatment of which begins with
equivocity (21,15–22,1).
If this interpretation is correct, then Andronicus’ influence on the
future course of Aristotelian philosophy was profound. The appear-
ance of the previously obscure treatise Before the Topics at the front of
an important new catalogue and publication of Aristotle’s esoteric
treatises drew attention from different philosophers, who interpreted
the treatise according to their own framework and interests. An
important example is Eudorus of Alexandria, who developed similar
moves in the interpretation of the text, bringing the ten categories

30
For examples and discussion, see Bobzien 2005.
31
Thus Porphyry delineates ‘predication’ or ŒÆŪæÆ in the course of a defence
of the title which, as I argue, might be traced back to Andronicus’ arguments against
the alternative title Before the Topics. In defending this interpretation, Andronicus
anticipated his pupil Boethus’ influential interpretation of the Categories as a treatise
about semantics (‘çøH ÅÆØŒH ± ºH ŒÆŁe ÅÆØŒÆ NØ H æƪø’
Porphyry in Cat. 58,5–6); and since fallacies of equivocation belonged to a wider class
of ambiguities that had seemed to some to threaten to derail, on sceptical grounds, the
power attributed by Antiochus to dialectic (cf. Cic. Ac. 2.92–6 = LS 37H), Andronicus’
emphasis on the Categories, and its rules for successful predication, might also have
taken aim at one weapon in the sceptical arsenal that had been deployed against
Antiochus. Within a few generations, at least, we begin to notice the ‘ten categories’
playing a basic role in defenses of the value and methodology of division, definition,
and demonstration, not only in Peripatetic sources but also among Platonists (see for
example Plutarch De Proc. An. 1023D–F, Alcinous, Didasc. 6, 159,43–160,30).
32
See Bodéüs 2001: xiv–xvi for Andronicus’ potential reliance on Stoic sources for
the progression of terms, statements, syllogisms, and demonstrations.
12 This Study
under the Platonic bifurcation of ‘absolute’ and ‘relative’ (Sophist
255D), and praising the practice of division ( ØÆæØ) as useful for
gaining knowledge—although Eudorus contrasted Aristotle’s (or An-
dronicus’ Aristotle’s) interpretation of the ten categories unfavour-
ably to a different interpretation, which he might have drawn
independently from the Old Academy. But there was an important
difference. Andronicus had found methodological and pedagogical
value in the Categories based on a more or less Stoic pedagogical and
psychological framework, since it was helpful for refining untutored
preconceptions or concepts, leading ultimately to scientific demon-
stration. Eudorus, by contrast, read the Categories ontologically, as a
direct map to the genuine tenfold structure of reality (perhaps, as
I tentatively suggest in ch. 3, to both sensible and intelligible reality).
In so doing, Eudorus inspired an influential Pythagorean reinterpret-
ation of the work (by the author we now call Pseudo-Archytas), but
also stimulated criticism from other Platonists (in particular the
shadowy figure Lucius, sometime in the first century bce or ce,
who was later followed in the main by Nicostratus and Plotinus).
In the next generation, Boethus of Sidon elaborated the interpret-
ation of the Categories that I have attributed to Andronicus, who was
probably a teacher or associate of Boethus.33 Boethus agreed that the
treatise was appropriate to novices, whose æ ºÅłØ of the genera of
predication required articulation,34 and he also maintained that the
treatise offers helpful ‘delineations’ or hupographai of the genera,
sketches that are able to ‘stir up’ our concepts (Simplicius in
Cat. 159,10–15) and begin to articulate them. But Boethus refined
Andronicus’ view considerably in response to a growing chorus of
Platonist critics of Aristotle’s Categories35—or more precisely, of
Andronicus’ version of Aristotle’s categorialism. Boethus stressed
that the treatise dealt strictly with sensible being, because it offered
an account of significant verbal expressions that signify (in the first

33
I treat Boethus as Andronicus’ pupil, on the strength not only of the biographical
tradition (e.g. Philop. in Cat. 5,19) but also based on his adaptation and revision of
Andronicus’ ideas, both in the Categories and elsewhere (e.g. Aspasius in Eth. 44,24).
34
This line of argument is attributed to the Peripatetic Herminus, in turn following
Boethus, at Porph. in Cat. 59,17–33.
35
Such as those attributed to ‘Lucius’ in Simplicius. For Boethus as a respondent
to Andronican challenges to the positions of Aristotle in Categories, see Moraux
1973: 147.
 Introduction 13
imposition) sensible things (Simplic. in Cat. 78,4–9): in Boethus’
formulation, the book is about ‘simple words significant of simple
things, qua significant’ (cf. Porph. in Cat. 58,3–7). Boethus contended
that these referents must be perceptible (ÆNŁÅ), so that some of the
criticisms offered by Lucius, which have to do with intelligible
beings ( ÅÆ), are irrelevant. Boethus’ formulation also would have
addressed (possibly earlier) ‘linguistic’ criticisms levelled by the Stoic
Athenodorus, which treated the Categories as dealing with verbal
expressions as such or º Ø ŒÆŁe º Ø (Porph. in Cat. 59,10–14
and Simplic. in Cat. 18,28–19,1), a view that Boethus seems to reject
explicitly; but there is not very strong evidence that he actually did
respond to Athenodorus by name or directly.
Boethus also developed, corrected, and improved Andronicus’
interpretations on many particular points, and introduced several of
the most important and penetrating questions about the Categories’
intent and scope, including his contrast of its account of substance
with that found in the central books of the Metaphysics; indeed
Boethus was prepared to consider the Categories a work of first
philosophy from a certain perspective, that is, insofar as it treated
real beings through the medium of words that successfully referred
to them. (He also may have rehabilitated the De Interpretatione and
with it a new, Aristotelian semantics that supported Andronicus’
view of the value of the Categories in terms of the sketch of a theory
of meaning found in the first chapter of De Int., creating a role of
concepts mediating between words and things: for this argument,
cf. Griffin 2012a).
In conclusion, I briefly suggest how Boethus’ view prevailed in
later antiquity, among teachers of Alexander of Aphrodisias like
Herminus, how it influenced Porphyry, and how later Neoplatonists
entertained a position that Eudorus may have favoured, that the ten
categories did offer a map of intelligible being and not merely of
sensible being.36 Eudorus took the view, I think, that the ten cat-
egories were correctly applicable to reality, but Aristotle’s description
of them (presumably in Cat.) was in some respects inadequate;
see ch. 3.

36
For the Neoplatonist discussion from Plotinus onward, see for example van den
Berg 2008 and Chiaradonna 2002.
14 This Study
Goals, Methods, and Materials
There are several merits in such a close study of the testimonies to
the reading of the Categories belonging to this era. One is the recon-
stitution of fragmentary evidence, which has not previously been
comprehensively collected and analysed: the more time is spent
combing through the later sources for (sometimes implicit) testimonia
and imitations, the more information about the lost works might be
obtained. Another is the influential nature of these early interpret-
ations. At least as the later tradition beginning with Porphyry repre-
sents them, the commentators of the first and second centuries ce
established the boundaries of how the Categories might be read, why it
might be important, and what function it served within Aristotle’s
philosophy. Many of their ideas and arguments have been recapitulated
throughout the tradition of subsequent commentary on the Categories,
and recent scholarship is no exception. Moreover, there may be some
room to point out places where the preconceptions of the first-century
commentators, once these are identified, have continued to shape
the readings of the text until the present. While such broad implica-
tions are partly outside the scope of this study, it is hoped that
the analysis here may provide a useful basis for further research
along these lines.
This study proceeds through a close study of the fragments and
testimonia for several of the key personae spanning the period
under consideration, in particular: Andronicus of Rhodes, Boethus
of Sidon, Lucius, Nicostratus, Athenodorus, and L. Annaeus Cornu-
tus. A briefer treatment is offered for several other personae, in
particular: Eudorus of Alexandria, Ariston of Alexandria, ‘Archytas’,
Aspasius, Adrastus, Taurus, Herminus, and Sosigenes. Several other
Peripatetics, Stoics, and Platonists are treated in passing. The histor-
ical figures under consideration are listed fully below (‘Personae’),
and discussed in detail throughout the main body of the study.
The majority of the important philosophical fragments are drawn
from the later Neoplatonic commentaries on the Categories in Greek:
in particular, from Porphyry’s shorter surviving commentary, from
Dexippus’ commentary (which, according to Simplicius, is effectively
a selective rendition of two lost commentaries, namely Porphyry’s
longer ad Gedalium and Iamblichus in Cat.), from Simplicius’ own
commentary, and from the later Alexandrian commentaries, mainly
those of Ammonius, Olympiodorus, and David (Elias). Simplicius’
 Introduction 15
voluminous surviving commentary is the backbone of these frag-
ments, and often provides attributions where the others lack them.
I have used a number of other sources in Greek for philosophical
fragments and testimonia—Plutarch, Aspasius, and Plotinus, for
example—but most of the earlier Greek texts, such as Strabo and
Lucian, have been deployed for historical and biographical detail. In
Latin, Boethius’ commentaries have also proved to be a useful philo-
sophical source, but most of the other Latin sources found in the
Appendix—such as Cicero and Persius—have been deployed here
just for historical and biographical detail. I have occasionally been
able to draw on Arabic sources, as in the chapters on Andronicus and
Herminus; in these cases, I have been reliant on editors’ translations
and previous commentators. At time of writing, I was not able to
consult in detail the exciting new commentary on the Categories
found in the Archimedes Palimpsest (Chiaradonna, Rashed, and
Sedley 2013), although I have been able to make reference to a few
supporting passages from the text.
Secondary literature is cited separately for each chapter and col-
lected in the bibliography. Like all students of the Peripatetic tradition
in the early empire, I am frequently indebted to Paul Moraux’s
foundational Aristotelismus I (1973) and II (1984), even (or espe-
cially) where I have ventured an alternative interpretation of the
scarce evidence. Among more recent surveys of the period in ques-
tion, I am especially reliant on those by Gottschalk (1987) and
Sharples (2008, 2010). Schofield (2013) is an outstanding new con-
tribution to the philosophical developments of the first century,
which I have found very valuable in correcting and developing ideas
developed in my earlier work (Griffin 2009). For the collection of
Neoplatonic sources on Aristotle, I have made liberal use of the three-
volume Sourcebook recently published by Richard Sorabji (2005),
especially volume 3, for subjects as distant from each other (at least
to the modern eye) as homonymy and Iamblichean ‘noeric exegesis’.
With respect to the Neoplatonic approach to Aristotle, I have also
appreciated the perspectives of Ilsetraut Hadot (e.g., 1991) and Rainer
Thiel (2004). More generally, I have been fortunate to rely on a
number of penetrating essays in two recent survey volumes edited
by Richard Sorabji and Robert Sharples (2007), among which I may
mention in particular those by Barnes, Reinhardt, Sorabji, and Shar-
ples, and in the earlier volume Aristotle Transformed, also edited by
Richard Sorabji (1990). On subjects of Middle Platonism, I have
16 The Structure of the Categories
gained a great deal from the classic treatment of Dillon (1977, repr.
1996) and the more recent work of George Karamanolis (2006).
Where the fragments have demanded the traversal of difficult
conceptual terrain, I have found a number of guides very useful;
these are cited in the footnotes and bibliography as they occur, but
here I can mention papers by Barnes and the seminal work of Patzig
translated by him (for passages relating to the syllogism), several
classic texts by Sorabji such as Time, Creation, and the Continuum
(1983), Marwan Rashed’s innovative Essentialisme (2007), and mul-
tiple articles on the Hellenistic theories of language in the volume
Language and Learning (Frede and Inwood 2005). I have also found
useful materials in the Australian ‘Reception of Categories’ project,37
and I am grateful to Professor Harold Tarrant for drawing these
materials to my attention, as well as sharing his own valuable survey
(2008). Last but by no means least, I have profited greatly from the
tremendous array of close commentaries and translations of nearly all
of the source texts in the ever-growing Ancient Commentators on
Aristotle series now published by Bloomsbury Academic. Unless
otherwise cited, translations from the commentators are from the
Ancient Commentators series.

II. THE STRUCTURE OF THE CATEGORIES

A short refresher on the content of the treatise Categories may be

helpful before beginning this study of its reception. (See Appendix 3
for a fuller survey.) The Categories was divided into three sections,
probably prior to Andronicus’ publication,38 and perhaps by an
editor who compiled scraps of text from different places to create
our treatise (cf. Frede 1987: 11–17):
(1) The first section is traditionally called the ante-praedicamenta
(‘what comes before the predicates’). It includes (ch. 1) a
description of three linguistic phenomena called homonymy
(equivocity), synonymy (univocity), and paronymy, followed

37
Accessed online at <http://categories.org.au>.
38
Since Andronicus protested the attachment of the postpraedicamenta, and
defended the inclusion of the antepraedicamenta. See below.
 Introduction 17
by (ch. 2) a twofold division of things said (ºª Æ), based on
whether they are said ‘without combination’ (nouns or verbs
interpreted as significant in isolation) or ‘with combination’
(sentences). Next (ch. 3) comes a celebrated, fourfold division
of beings (ZÆ) determined by their role in the predicate
position of a sentence, where, roughly, they may either name
a kind to which the subject belongs (in Aristotle’s language, be
said-of the subject), or name some property that inheres in
the subject but is not its kind (be in the subject, in a certain
sense that Aristotle attempts to define technically). This sec-
tion sparked many ancient debates and fruitful lines of inquiry.
For example, (a) in ch. 2, the division of ‘things said’ and
‘things that are’ helped to fuel a debate over the proper subject
of the Categories—words, concepts, or things?—which culmin-
ated in a compromise solution offered by Boethus of Sidon: the
treatise concerns semantics, how words signify things. And (b)
Andronicus of Rhodes, as I argue below (my ch. 2) found
much use in the distinction between items said-of and in a
subject, and leveraged this distinction to reinvigorate Aristo-
telian essentialism. This chapter, after all, suggests that there is
something essential about kinds, and that one might be able to
use predicative practices in natural languages as a kind of
heuristic guide for recognizing the essences of things.
(2) The second section is traditionally called the praedicamenta
(‘predicates’ or ‘categories’). It includes (ch. 4) a list of ten
‘things said without combination’, roughly translatable as
reality (for example, man, horse), quantity (four-foot,
five-foot), quality (white), relative (double, half), where
(in the marketplace), when (yesterday, last year), in-position
(lying down), having (shod, armed), acting (cutting, burn-
ing), and being-acted-on (being cut, being burned): this has
struck many readers as a list of the most universal, or ‘ultim-
ate’, genera of being (a list which then ought to be distin-
guished carefully from the ten types of predicate that are
sketched in Topics 1.9, with Frede 1987: 35–6, although our
commentators were perhaps less sensitive to this distinction).
The list is then followed by a sketch of distinguishing features
of some of these, namely (ch. 5) reality, (ch. 6) quantity,
(ch. 7) relative (appearing out of order, compared to ch. 4),
18 The Structure of the Categories
(ch. 8) quality, (ch. 9) acting and being-acted-on. These
‘sketch accounts’ or ªæÆçÆ were (as I argue below, espe-
cially in chs. 2 and 5) understood by some ancient Peripatetics,
like Andronicus, Boethus, and Herminus, as valuable tools
used to ‘stir up’ our preconceptions ( æºłØ) of the highest
genera, and therefore are especially useful for beginners to
philosophy: this argument may have underwritten the case
for the Categories as a novice’s introduction to the subject of
philosophy. Be that as it may, Aristotle’s examinations of these
individual categories attracted swathes of discussion in their
own right, now largely preserved in Simplicius’ commentary;
Boethus of Sidon, for instance, composed a whole book On the
Relative. The praedicamenta conclude with a later interpol-
ation briefly filling in the remaining categories and explaining
that not much needs to be said about them. Perhaps the
original manuscript trailed off here, and an editor felt com-
pelled to insert a short passage explaining why.
(3) The third section is called the post-praedicamenta (‘what
comes after the predicates’), and includes (chs 10–15) relatively
disconnected discussions of opposition, priority, simultaneity,
movement, and the usage of the word ‘have’. Andronicus, as we
will notice in ch. 2, treated the first two parts as a coherent
and interdependent whole, but maintained that a later editor,
who was confused, had added the final section to the treatise
in error.
Appendix 3 outlines these sections in more detail. For a fuller
treatment of the Categories itself readers may refer to a number of
excellent introductions (including Bodéüs 2001: xi–clxxviii, Wedin
2000, and Ackrill 1963).
 Part A

Rediscovery and Endorsement:

Andronicus and Eudorus
 2

Andronicus of Rhodes

In this chapter, I argue that a late ancient interpretation of the

Categories can be traced to Andronicus of Rhodes in the first century
bce. According to this interpretation, the treatise helps to train us in
the demonstration of truth (IØØ), and not merely in the practice
of persuasive rhetoric or dialectic.1 Andronicus sought to breathe new
life into Aristotle’s vision of demonstrative science, and he found the
Categories especially valuable for this function. In particular, Andro-
nicus believed that the Categories helps us to distinguish per se
predications whose subject and predicate both fall in the first category
(for instance: ‘Socrates is a man’) from per accidens predications
whose predicate falls in one of the non-substance categories (for
instance: ‘Socrates is pale’, ‘tall’, or ‘hungry’).2 The former, Androni-
cus argues, can be used to construct good definitions and so to engage
in the rudiments of demonstration by way of division (ØÆæ Ø),
while the latter cannot. This interpretation partially explains Andro-
nicus’ motivation for placing the treatise in so prominent a position
in his catalogue of the Aristotelian corpus and rechristening it,

1
The dichotomy between demonstration and persuasion was, I think, treated as
exhaustive by Andronicus, following the Hellenistic tradition (compare Diogenes
Laertius 5.28, discussed below: rhetoric and dialectic are treated as proper subdivi-
sions of persuasion, and both as distinct from demonstration of the truth). If my
interpretation is correct, Andronicus treats the Topics as concerned with the latter,
and used the word ‘dialectic’ in the sense of DL 5.28, as a subdivision of persuasion,
not in the Stoic sense, where it might include demonstrative proofs.
2
Based on the examples that I draw from Boethius De Div. below, I think this
argument was made primarily about the definition of substances (for example, ‘man is
an animal’); defining non-substances (‘black is a colour’) seems to have been regarded
as less interesting by Andronicus, or at least is less represented in the fragmentary
sources. As I also discuss below, Andronicus seems to have treated the first category as
(more or less) a category of substance ( P Æ).
22 Andronicus of Rhodes
promoting the title Categories in place of the apparently earlier title
Before the Topics.
I suggest that Andronicus found the Categories to be suitable for
beginners in philosophy, primarily due to the relatively non-technical
manner in which Aristotle wrote about the categories themselves—
the central descriptions of the features of substances or real beings
(ch. 5), quantities (ch. 6), relatives (ch. 7), qualities (ch. 8), and several
other non-substance categories. I argue further that on Andronicus’
view, the descriptions of the ten ŒÆ Ū æÆØ in the central chapters of
the Categories articulate the beginner’s innate preconceptions
(æ º łØ) of the most general kinds of being. The Categories is
particularly effective in this introductory capacity because it offers
only  ªæÆçÆ, ‘sketch accounts’ or ‘delineations’, of each genus
rather than attempting to provide formal definitions (which would
be difficult for the beginner, and impossible to provide for the highest
genera). Thus pedagogical considerations, underwritten by a more or
less Hellenistic epistemology, also underwrite Andronicus’ value for
the treatise. In addition, there is some pragmatic value in the Cat-
egories for use in debate: its  ªæÆçÆ provide simple heuristics for
checking the genus to which a given term belongs, enabling us to
defeat various sophistical arguments that turn on equivocation by
distinguishing clearly between essential and accidental modes of
predication.
This way of exploiting the Categories, I think, helps to explain why
Andronicus found the work interesting and relocated it to the front of
his influential catalogue and publication of Aristotle’s treatises. That
relocation in turn drew attention to the work and spurred wider
interest in it. I do not want to argue that Andronicus’ substantive
ideas about the treatise were themselves the source of that interest, but
I do suggest that there are signs of engagement with Andronicus’
thought in the contemporary and later exegetical tradition on the
Categories. His interpretation may have been followed in the main
by Boethus of Sidon (see ch. 6), and in turn influenced Porphyry and
the later Neoplatonist tradition. I also argue in ch. 3 that there are
signs of philosophical engagement with Andronicus in the evidence
for Eudorus’ interpretation of the Categories (and perhaps vice versa).
And even if Andronicus did not think that the Categories was a work
of ontology, his emphasis on the distinction that it drew—on his
view—between talk of real beings (picked out through per se predi-
cation) and their accidents (picked out only in their relation to other
 Andronicus of Rhodes 23
things) anticipates the ontological reading of the Categories that
would become common even to the present.

I. LIFE AND HISTORY

@æ ’ Kª   Å ¼Ø Ø  ºº d æÆ Åº ÆØ  ŒÆd

IŁºÅ Æ, z N Ø ŒÆd ƒ —ÆÆØ  ı F çغ ç ı æª  Ø·
H b  ºØ ØŒH ŒÆd H æd ºª ı ŒÆd çغ çÆ ‹ 
—ÆÆ Ø  ÆP e ŒÆd  æÆ ŒºB ŒÆd æØŒ  › KŒ H
æØ ø ŒÆd ¸øÅ › øØŒ.
Many men worthy of remembrance, commanders and athletes,
were born [on Rhodes]. Among them are the forebears of
Panaetius the philosopher. Among the statesmen, rhetoricians,
and philosophers are Panaetius himself; Stratocles; Andronicus
the Peripatetic; and Leonides the Stoic.
Strabo 14.2.13
Regrettably, we remember relatively little of Andronicus the Peripat-
etic, Icæ  Å ¼Ø .3 From Strabo we know that he was born on

3
For a review of Andronican studies for the better part of the twentieth century,
see in particular P. Moraux, Aristotelismus I (Berlin, 1973), 45–141 and J. Barnes,
‘Roman Aristotle’, Philosophia Togata II (Oxford, 1997), 1–69. Now see also Sharples
(2008), Hatzimichali (2013), and to Barnes, contrast Primavesi, ‘Ein Blick in den
Stollen von Skepsis: Vier Kaptiel zur frühen Überlieferung des Corpus Aristotelicum’
(2007). Formative treatments include F. Littig, Andronicus von Rhodos, 3 vols
(Munich, 1890; Erlangen, 1894, 1895) and M. Plezia, De Andronici Rhodii studiis
Aristotelicis (Cracow, 1946). See also E. Zeller, Die Philosophie der Griechen in ihrer
Geschichtlichen Entwicklung, T. III: Die Nacharistotelische Philosophie (Fotomech,
Nachdr. der 5. Aufl., Leipzig 1923), 1, 642 and following; K. O. Brink, ‘Peripatos’,
Paulys Realencyclopädie, suppl. 7 (1940), cols. 899–949; I. Düring, Aristotle in the
Ancient Biographical Tradition (Göteberg, 1957), 412–25; F. Wehrli, Die Schule des
Aristoteles (2nd ed. Basel, 1967–9); H. B. Gottschalk, ‘Aristotelian Philosophy in the
Roman World from the Time of Cicero to the End of the Second Century ad’, ANRW
II.36.2 (1987), 1079–174, esp. 1097–120; R. Goulet in R. Goulet (ed.), Dictionnaire des
Philosophes Antiques (Paris, 1989), 200–2; R. W. Sharples, ‘The Peripatetic School’, in
D. J. Furley (ed.), Routledge History of Philosophy, vol 2: From Aristotle to Augustine
(London, 1997), 147–87; and L. Taràn’s significant review of Moraux, ‘Aristotelianism
in the 1st century bc’, in L. Taràn, Collected Papers (Leiden, 2001), 479–524. On the
relationship of Andronicus De Div. to Boethius De Div., see also G. Pfligersdorffer,
‘Andronikos von Rhodos und die Postprädikamente bei Boethius’, Vig. Christ. 7
(1953), 98–115 and J. Shiel, ‘Boethius and Andronicos of Rhodes’, Vig. Christ. 11
(1957), 179–85. On the work de passionibus ascribed to A., see A. Glibert-Thirry,
Pseudo-Andronicus de Rhodes: —æd ÆŁH = Corpus latinum commentariorum in
Aristotelem graecorum, suppl. 2 (Leiden, 1977).
24 Life and History
Rhodes, a centre of Peripatetic learning since Eudemus; his later
eponym ›   Ø  (Plutarch, Sulla 26.1, 9) is consistent with this origin,
and his philosophical affiliation is suggested in epithets such as › KŒ
H æØ ø or æØÆ Å ØŒ by Strabo, Galen (An. Corp. 4.782
Kühn), and Porphyry (VP 24,7). His work was sufficiently familiar to
later Peripatetic commentators that he could be mentioned as
æØŒ  without prior introduction or qualification (Alexander in
An. Pr. 161,1; Porphyry in Cat. 125,22), and his lucidity as an exegete
of the Aristotelian text was respected in late antiquity, particularly
following Porphyry: he wins praise from Themistius (in De. An.
32,22–4), and Ammonius refers to him as a successor to Aristotle in
the Peripatetic scholarchate (in De Int. 5,29).4 The tradition about his
scholarchate seems to presuppose a degree of respect for Andronicus
as a Peripatetic philosopher, either from his contemporaries (if the
testimony is literally accurate) or from Ammonius’ later sources (if
the testimony was invented, or a product of guesswork). The sources
for Andronicus’ life and thought extend from the first century bce to
the sixth century ce, with their focus shifting from (1) biography and
bibliography with Strabo and Plutarch (I bce–I ce)5 to (2) psychology
with Galen, Aspasius, and Themistius (I–IV ce)6 before coming to

4
But the credibility of Ammonius’ testimony has been doubted; elsewhere, he
bestows the same number (‘eleventh [scholarch] after Aristotle’) on Boethus of Sidon
(in An. Pr. 31,11): cf. Lynch (1972), 203–4 and Brink (1940), 938–40.
5
Strabo, the only contemporary source, mentions Andronicus as a ‘noteworthy’
Rhodian philosopher (14.2.13). A century later, Plutarch (Sulla 26.1–2) remarks on
the reported role of Andronicus in the recovery and organization of the Aristotelian
texts, completing the tale curiously left unfinished in (our text of ) Strabo’s thirteenth
book (13.1.64). Later, Porphyry (VP 24.7) is clearly interested in Andronicus as a
bibliographer of Aristotle, and renews Strabo’s and Plutarch’s intrigue with the fate of
the Aristotelian and Theophrastan library and Andronicus’ pinakes.
6
Galen (Quod animi mores 782,14) refers to an Andronican account of the soul as
a ‘blend’ (ŒæA Ø) or ‘power’ of the body: as we shall see, this doctrine is plainly
expressed in a Stoic context. The Christian apologist Hippolytus (if he is the author of
Ref. 5.21.1–5) curiously notes Andronicus’ doctrine of ‘blending’ as an authoritative
resource for the Gnostic Sethians in describing the relationship of soul to matter.
Aspasius (in Eth. 44,18–45,5) focuses on Andronicus’ account of pathos as a hupo-
lēpsis of the soul, echoed by the pseudo-Andronican Peri Pathōn, itself a highly
Stoicizing work. In the fourth century, Themistius (in An. 31,1–32,34) provides a
first-hand quotation from Andronicus in defence of Xenocrates’ doctrine of the soul
as a self-moving number. In this full paragraph Andronicus describes psychic ŒæA Ø
in his own words, explaining that ‘the [self-moving] soul is the cause of the blend’ of
‘the primary elements in accordance with specific ratios and numbers’.
 Andronicus of Rhodes 25
rest upon (3) logic with Porphyry7 and the later Neoplatonists
(III–VI ce).8
Andronicus is  Å ¼Ø  because his name is attached to a
prodigious expansion in the study of Aristotle’s ‘esoteric’ texts during
the first century bce, whose causes are not yet clearly understood.9
(The label ‘esoteric’ applies to the technical treatises by Aristotle that
we now possess, such as the Topics and Physics; it is used to contrast
these treatises against Aristotle’s ‘exoteric’, published Dialogues,
which were popular in antiquity but have survived only in fragments.
Andronicus himself, in drawing attention to a series of letters
attributed to Aristotle, may have done something to popularize the
distinction and privilege the esoteric texts).10 Andronicus’ attachment

7
Alexander of Aphrodisias (in An. Pr. I 160,28–161,1), with later sources such as
Ammonius (in De Int. 5,28 and 7,13) and Philoponus (in Cat. 27,18–26 and 45,8–12),
takes an interest in Andronicus’ opinion, evidently unique in antiquity, that the De
Interpretatione was not written by Aristotle. Porphyry famously develops an abiding
interest in the Categories, and in this he seems to have taken his cue from Andronicus.
Indeed Andronicus is quoted in Porphyry’s commentary by Question and Answer
(125,22); and Porphyry’s full Commentary on the Categories addressed to Gedalius
must have contained many more references, to judge by the array of named citations
in Simplicius (including in Cat. 21,21–4; 26,17; 30,3; 54,8–21; 63,22–8; 134,5;
142,34–143,1; 144,7–15; 150,31–151,7; 153,29–155,2; 157,18–22; 159,32; 202,5;
214,22; 258,15; 163,19–22; 266,6; 269,21; 270,2; 332,15; 342,23; 347,6; 357,28;
359,16; 379,9–11; 385,3–9). The De Divisione of Boethius, which draws on an essay
by Porphyry in his Sophist commentary, cites Andronicus and may be modelled on an
Andronican publication of the same name (Andronici . . . de diuisione liber editus,
4,3–11 Magee).
8
We know that Iamblichus followed Porphyry’s lead in commenting on the
Categories, and a notice in Simplicius indicates that Iamblichus ‘followed Andronicus’
in at least one point of exegesis (in Cat. 144,7–14). Simplicius himself offers the most
prolific preserved resource for Andronican doxography in his own Commentary on
the Categories; Simplicius appears to be heavily dependent on Porphyry, and the
names of Andronicus and Boethus do not frequently occur except in Porphyry’s
company. The later Neoplatonists, including John Philoponus (in Cat. 5,19) and
Ammonius, offer some biographical details that have been doubted due to their
late date, including Andronicus’ putative scholarchate and his relationship to
Boethus of Sidon.
9
See for example Moraux (1973) I; Gottschalk (1987), 1089–97; Tarán (1981);
and Sharples (2008).
10
On the distinction between exoteric works and esoteric (or ‘acroamatic’) works,
and their respective readership in the Hellenistic period, see for example Barnes 1997:
12–16, Hatzimichali 2013: 12, 26. There is already some evidence in Aristotle for
the distinction, for instance in Eudemian Ethics 2.1 1218b34 (‘All goods are either
outside or in the soul, and of these those in the soul are more desirable; this distinction
we make even in our popular discussions’). On the role of Andronicus’ publication of
the ‘letters’ in helping to cement the distinction and privilege the esoteric works, see
26 Life and History
to the expansion of interest in Aristotle’s ‘esoteric’ treatises is first
explicitly attested, in our record, by Plutarch (Sulla 26); a century
earlier, Strabo, while speaking in markedly similar terms (Geog.
13.1.54),11 describes the sorry state of Aristotle’s works in the hands
of inferior booksellers, but stops short of mentioning the editorial
rescue allegedly mounted by Andronicus12—whom Plutarch credits
with a corrected, usable publication of Aristotle and Theophrastus.
This collection served to revitalize a Peripatos previously incapaci-
tated, on Strabo’s account, by the loss of its founder’s school texts.
Coupled with the report of Porphyry (VP 24) that Andronicus organ-
ized the corpus into its contemporary form, these texts have trad-
itionally substantiated the legend of a Roman ‘critical edition’ woven
in Andronicus’ hands, constituting the foundation of our modern
Aristotle. The evidence for this tradition is scanty,13 especially if we
understand textual criticism to be the central activity of an editor;14
but if we understand the cataloguing and organization of the Aristo-
telian canon as Andronicus’ primary achievement, his claim to fame
seems valid. This leads to the central point that I hope to highlight in
the following sections: whatever the status of Andronicus’ editorial
achievement, his organization of the reading order of the Aristotelian
treatises had a lasting impact, as Plutarch (Sulla 26.1–2) and Por-
phyry (Life of Plotinus 24.7) both attest. Andronicus’ relocation of a
(then little-known) Aristotelian treatise to the forefront of his cata-
logue deserves to be treated as an influential event in the history of
philosophy.

Hatzimichali 2013: 26; Aulus Gellius (Attic Nights 20.5) links Andronicus to the
correspondence attributed to Aristotle. I am grateful to Stephen Menn for comments
on an earlier draft that helped to clarify this issue.
11
On the possibility of a common source, see Moraux I: 21–4, with Barnes 1999: 9,
19–20.
12
Perhaps, as Barnes suggests, because the text is corrupt. In that case, Plutarch
may be simply reporting Strabo. On the other hand, H. Diels, Doxographi Graeci
(Berlin, 1879) 216 proposed that Plutarch simply fabricated a connection between
Strabo’s Apellicon story and the fact that the current catalogue of the day carried the
name of Andronicus. See Moraux I: 48 and following for discussion.
13
See Barnes (1997), with discussion below.
14
For the division of ‘editorial’ work into textual criticism and canon-organiza-
tion, see Hatzimichali 2013:1, 18–23. Note for example that Andronicus did not read
ºª  B P Æ in the opening lines of the Categories (in Cat. 54,8–21: his text omits
B P Æ, perhaps in keeping with his wider point that predication of a subject is not
limited to the first category), and was not followed in the later tradition.
 Andronicus of Rhodes 27
When did Andronicus live and work? This is a knotty problem. It is
widely agreed that he was active in the first century bce. According to
John Philoponus (in Cat. 5,16–19), Andronicus taught Boethus of
Sidon, himself a ‘brilliant’ scholar (ŁÆı Ø , Simplicius in Cat. 1,18)
whose ‘quick-witted’ ( ººB IªåØ Æ, 11,23) defences and exegeses
of Aristotle’s Categories might appear, to a loyal reader of Simplicius,
to have left his mysterious Platonizing opponent Lucius intellectually
far behind. If we trust Philoponus, we can set a reasonably sure date
for Andronicus’ activity. Boethus ‘philosophized with’ Strabo
( ıçغ ç Æ, Geog. 16.2.24) as a teacher or a fellow-student,15
and Strabo’s birth can be fixed around 63/64 bce.16 That might place
Boethus’ activity, as a student of Andronicus, not much before the
middle of the first century bce, and consequently establish Androni-
cus’ floruit around the late seventies (Moraux)17 or sixties
(Gottschalk).18 (Boethus’ plain adaptation and defence of Androni-
cus’ definition of pathos (Aspasius in Eth. 44,24) lends weight, I think,
to the biographical claim that he was Andronicus’ pupil.)
We might be reluctant to rely upon Philoponus’ isolated notice,
however, which was composed centuries after the fact.19 Düring,
following Brink,20 would view Andronicus and Boethus as near
contemporaries, and date Andronicus’ publication of Aristotle and
Theophrastus around the thirties bce: after all, Cicero would surely
have mentioned Andronicus had a seminal Peripatetic publication
circulated under his name before 43 bce, yet Cicero calls another
man, Cratippus of Pergamon,21 the leading Peripatetic in Athens (Off.

15
The ambiguity turns on the force of ıçغ ç Æ + dat. in fiz
ıçغ ç Æ E a æØ ºØÆ; it could imply ‘as a teacher’, or ‘as a
fellow-learner’.
16
On Strabo’s dates, see W. Aly, ‘Strabo (3)’, Realencyklopädie der classischen
Altertumswissenschaft 4 (1932), 76–155.
17
P. Moraux, Der Aristotelismus bei den Griechen, vol. 1, Berlin 1973, pp. 45–55.
18
H. B. Gottshalk, ‘Aristotelian Philosophy in the Roman World from the Time of
Cicero to the End of the Second Century ad’, in H. Temporini and W. Haase, Aufstieg
und Niedergang der römischen Welt, vol. 36. 2, Berlin 1987, pp. 1079–174, in particu-
lar pp. 1095–6.
19
It has been argued that the later Neoplatonists, such as Philoponus and Ammo-
nius, really knew nothing about Andronicus, save what they read in the sources
available to us: see L. Tarán, ‘Aristotelianism in the 1st century bc’, in Tarán, Collected
Papers, Leiden/Köln/Boston 2001, pp. 479–524, in particular pp. 495–7.
20
Düring (1957), 420–5 proposes a date between 40 and 20 bce; for earlier views
see C.O. Brink in RE suppl. 7 (1940), 938.
21
On Cratippus see Moraux I: 223–56.
28 Life and History
1.1.1, 3.2.5; Tim. 1.1). Conversely, if Barnes (1997) is right to doubt
whether Andronicus’ work made any ‘splash’ at all during his life-
time, this line of reasoning loses its speculative bite; and Gottschalk
(1987) casts doubt on it for other reasons. We shall return to this
problem in more detail below.22
Where did Andronicus do his research, after an early life presum-
ably (on Strabo’s testimony) occupied on Rhodes? If Ammonius of
Alexandria is telling an historical truth when he claims the eleventh
Peripatetic ‘scholarchate’ for Andronicus,23 then it is possible to infer,
with Gottschalk and others, that Andronicus’ activity centred around
Athens. Adopting the later chronology, it has been proposed that
Strabo, Andronicus, and Boethus attended the same lectures,24 and
even met as young men in Alexandria, where Strabo spent time
(2.3.5); perhaps it was there that Strabo heard Tyrannio at some
point in the middle forties (cf. 12.3.16). But this is pure speculation.
It had at one point seemed fairly clear that Andronicus had to visit
Italy in order to produce his collection, the putative ‘Roman edition’
of Aristotle. However, it is not self-evident from Plutarch’s text (Sulla
26) that Andronicus obtained physical access to the particular copies
in Sulla’s library, as Tyrannio did.
What we do learn from Plutarch’s source, speaking in oratio
obliqua, is that Andronicus had ‘ready access’ to the copies ( a

22
In this context (De Off. 1.1, 3.2) Cicero is praising a tutor who was ‘like a father’
to his son (cf. Ad Tironem xxi): even if word of Andronicus’ work had already reached
him at the time of this composition, and even if he viewed these improved texts as
philosophically important, why should Cicero retail the scholarly merits of Cratippus’
contemporaries or rivals here? There is no reason to suppose that one of the greatest
contemporary Peripatetics (Cicero, Tim 1.1), renowned as a teacher, would lose his
mystique as soon as booksellers replaced Tyrannio’s lousy text with a better edition.
The argument ex silentio is more persuasive: but I do not see that an Andronican
‘edition’ or ‘collection’ could not have been published before 43 bce just because
Cicero fails to mention it (despite his readiness to give names); if Triclinius’ ground-
breaking work on Greek tragedy and comedy, fifteen centuries later, is a fair compar-
andum, it could easily take decades until a critical mass of scholars—the sources of the
tradition reported by Strabo and Plutrach—recognized the importance of such a work
of close philology. There seem to be two variables in play: (1) when Andronicus made
his collection public, and (2) when the sources of the tradition reported by Strabo and
Plutarch resolved that it was a superior text, and worthy of primary attention. I return
to this point below.
23
On which see Brink 1940: 938–40; against the report, see Lynch 1972: 203–4.
Barnes (2007) discusses the issues without commitment, including Littig’s simple,
perhaps too simple, emendation at in An. Pr. 31,11.
24
Düring (1957), 413, condemned by Barnes (1997: 3) as a ‘garbled invention’.
 Andronicus of Rhodes 29
I ªæÆçÆ) from Tyrannio, and proceeded to publish them (put them
N  ), himself writing up (IƪæłÆØ) the catalogues current in
Plutarch’s day. This language might suggest, with room for uncer-
tainty, that Andronicus spent time in Italy; but these I ªæÆçÆ
might just as well have been sent to him elsewhere, say, in Athens
or even Rhodes. Following Gottschalk, one might imagine Androni-
cus making his copies in Italy, and elsewhere across the Aegean,
before retiring to Athens, where his research rightly won over Ariston
of Alexandria—who followed him on several points of interest in the
exegesis of the Categories—as well as Cratippus of Pergamon from the
Academy (although this should likely not be regarded as a traumatic
‘apostasy’ of any kind, but a natural development).25 On the other
hand, if Andronicus’ research was not understood as epoch-making
or even as especially important in his own lifetime, this story is
evidently less attractive. It is also important to note that the very
earliest signs of interest in Aristotle’s ‘esoteric’ treatises (discussed
above), including the antecedent efforts of Apellicon and Tyrannio,
appear to predate Andronicus’ activity.26

II . WORKS AND EDITORIAL ACTIVITY:

BEGINNING WITH THE CATEGORIES

A number of works have been attributed to Andronicus, including his

catalogue of the treatises of Aristotle and Theophrastus, a paraphrase
of the Categories, a book On Division (which was read by Plotinus and
Porphyry, and later on influenced Boethius through Porphyry), and a
spurious, surviving treatise —æd —ÆŁH (ed. Glibert-Thirry 1977).27
But scholarly interest naturally focuses on reports of his ‘Roman
edition’ of Aristotle. The available sources for this editorial activity
can be read with varying results. Jonathan Barnes’ scepticism about
the sources is surely healthy,28 and sounds a cautionary note about

25
See Blank 2007: 92 and Chiaradonna 2013: 39–40, with Moraux 1973: 225–6 and
Donini 1977: 247–8.
26
For this point see Hatzimichali 2013: 17–18.
27
For a brief review of these works with references, see Appendix 2.
28
J. Barnes, ‘Roman Aristotle’, Philosophia Togata II (Oxford, 1999), esp.
pp. 24–44.
30 Works and Editorial Activity
the preconceptions latent in the earlier consensus broadly shared
(with some differences) by Gottschalk, Düring, and Moraux.
Here I would like to focus on the evidence for the claim that
Andronicus relocated the Categories in his catalogue of Aristotle’s
works, and that he had philosophical reasons for doing so. (A more
detailed overview of the issues relating to the ‘library’ of Aristotle’s
esoteric works can be found in Appendix 2).29 For these purposes, we
can leave aside the attractive image of Andronicus as textual critic and
compositor of treatises, and focus on his status as a publisher and a
cataloguer, and most importantly organizer, of the corpus. The order
in which Andronicus presented the works, and the arguments that he
made in favour of that order, appear to have been among his most
lasting contributions to Peripatetic thought. He certainly published a
catalogue (ƌ) of Aristotle and Theophrastus. It was the source
of the catalogues current in Plutarch’s day, on the cusp of the second
century ( f F çæ  ı ÆŒÆ, Sulla 26.1–2), and it would
function as an influential model for Porphyry in the third century, as
discussed below. The catalogue of ‘Ptolemy the Unknown’30 shows
that a catalogue carrying Andronicus’ name also possessed ‘canonical’
status in Ptolemy’s own lifetime (whenever that might have been),
and that it was not like the Hellenistic catalogue preserved by Dioge-
nes Laertius (5.22–27)—which, notably, omits our Categories (under
that title) altogether.31
More than a century after Plutarch, Andronicus’ stylistic example
was followed by Porphyry—and presented, as the Life of Plotinus
(24.7) implies, in order to underwrite and justify Porphyry’s own
approach to the Enneads. Porphyry famously observed that he fol-
lowed Andronicus who ‘divided the works of Aristotle and Theo-
phrastus into treatises and collected related material together’ ( a

29
If indeed there was a single library: see Y.L. Too, The Idea of the Library in the
Ancient World (Oxford, 2010), pp. 25–31. Natali and Hutchinson (2013: 101–4) offers
a clear summary of the story of Aristotle’s library.
30
See the edition of the catalogue of Ptolemy by Hein (1985); I am indebted to
Stephen Menn for pointing me to this work. The name ‘Ptolemy the Unknown’ or ‘the
Stranger’ likely derives from the misunderstanding of — ºÆE    as
— ºÆE  ˛ ; see Rashed 2005: ccvii, Hatzimichali 2013: 19 n. 57. The catalogue
had at least five books, if not more (see item 97 in Düring 1957: 230); Littig 1894:
18–25 offers a reconstruction, and Plezia 1946: 16–35 studies it in detail, as does
Moraux 1973: 58–94. For the importance of Diogenes’ catalogue, and the fact that it is
more than a mere ‘library list’, see Hatzimichali 2013: 23–4.
31
See Moraux 1951: 131, Düring 1957: 69, and Gottschalk 1987: 1103.
 Andronicus of Rhodes 31
æØ º ı ŒÆd ¨ çæ ı N æƪÆ Æ ØEº a NŒÆ
 Ł Ø N ÆP e ıƪƪ). This does not mean that Androni-
cus invented treatises by pulling together previously unrelated mater-
ial and adding bridge passages, but that he pulled related treatises
together, and within each collection organized the treatises, likely in
such a way that easier or ‘lighter’ material came first (as Porphyry
characterized his own practice: Œ fi Å b KØ a NŒEÆ çæø
ıçæÅ Æ  f ŒÆd Ø æ Å E KºÆçæ æ Ø æ º Æ Ø,
24.14–16). This final point, I think, deserves somewhat greater
emphasis than it usually receives: giving Ø æ Å E
KºÆçæ æ Ø æ º Æ Ø, the starting position to the lighter mater-
ial, is also the reasoning that Porphyry endorses in explaining the
location of the Categories at the outset of the Aristotelian curriculum
(in Cat. 59,21–22), and the argument that Porphyry provides for
the introductory nature of his own Isagoge (1,8–9) as a precursor to
the study of categories (N c H Ææa æØ ºØ ŒÆ Ū æØH
ØÆ ŒÆºÆ, 1,2–3).
We can infer from later reports that Andronicus’ catalogue opened
with the Categories and the other books that comprise the Organon
(probably including De Interpretatione, although this may have car-
ried a mark of dubious authorship),32 and that here, or elsewhere,
Andronicus provided some justifications for the position of ‘logic’ at
the propylaea of Aristotelian studies (discussed below); this organ-
ization of the corpus was also to prove influential. Indeed, it was one
area where the later tradition, including Porphyry and the Neopla-
tonists, favoured Andronicus’ view over that of his ‘pupil’ Boethus (cf.
Philop. in Cat. 5,15–20), who recommended commencing the study
of Aristotle with physics. As we have at least some evidence that
Andronicus felt the need to argue for the usefulness and importance
of logic in his effort to build an Aristotelian ‘system’ based on the
corpus, we may infer that this was not taken for granted, and that
Andronicus played something of an original role in asserting (or
reasserting) the importance of logic in Peripateticism, and of

32
Andronicus might have included De Int. in his catalogue, and maintained its
usefulness, while ‘athetizing’ it as the work of another author. That is, he might
consistently have endorsed the views represented in De Int. as useful and integral to
the project of the Organon, while denying that Aristotle wrote it. In this context, it is
worth considering that Andronicus did not think that the truth was coextensive with
Aristotle’s written views: for instance, Andronicus appears to have disagreed with
Aristotle concerning the names and functions of the categories ‘where’ and ‘when’.
32 Andronicus on the Categories
Peripatetic logic in general. This would be supported by the evidence
that Andronicus paid special attention to the Categories, if this is not
merely an artifact of the later sources.33

III. ANDRONICUS ON THE CATEGORIES

Next is the third heading, where in Aristotle’s writings one

should start. Well, Boethus of Sidon says that one should always
start from the treatment of physics, since this is more familiar
and knowable for us, and one should start from the things that
are more clear and knowable. But his teacher, Andronicus of
Rhodes, examining [the issue] more exactly, said that one
should first begin with logic, which is concerned with demon-
stration. (Philoponus, in Cat. 5,15–20, tr. Sharples 2010, lightly
modified)34
When Andronicus and his pupil Boethus defended their respective
teaching routines, it was a commonplace that philosophical discourse
(› ŒÆ a çغ çÆ ºª ) fell naturally into three provinces: logic,
ethics, and physics.35 Antiochus of Ascalon had traced the tripartition
to the Old Academy in the fourth century bce (Cicero, Ac. 1.19; for
the attribution to Xenocrates, see also Sextus, Adv. Math. 7.16).36 But
its lasting influence was mediated by the Hellenistic Stoa,37 where the

33
We are aware that Andronicus provided a ‘paraphrase’ of the Categories (so
Porphyry via Simplicius: in Cat. 26,17, 30,3), and we are not directly aware of any
other such ‘paraphrases’; this may merely be an accident of preservation, but in other
cases where Andronican ‘readings’ are cited, such as the Physics (cf. Simplic. in Phys.
440,13), it appears that Andronicus has offered a reading of the Aristotelian text
rather than an interpretive paraphrase. This too might suggest some special concern
with logic.
34
æ  q KçB ŒçºÆØ  e Ł E ¼æå ŁÆØ H æØ ºØŒH
ıªªæÆ ø. BÅŁ  b s çÅ Ø › ØØ  E Ie B çı ØŒB ¼æå ŁÆØ
æƪÆ Æ –  E ıÅŁ æÆ ŒÆd ªøæ ı, E b Id Ie H Æç æø
¼æå ŁÆØ ŒÆd ªøæø. › b  ı Ø ŒÆº  æØŒ  ›   Ø  IŒæØ æ 
K Çø ºª åæBÆØ æ æ  Ie B º ªØŒB ¼æå ŁÆØ, X Ø æd c IØØ
ŒÆ ƪ ÆØ (ed. Busse 1898).
35
See for example Ierodiakonou 1993. Aristotle divides ethical, physical, and
logical propositions at Topics 1.14 (105b19 ff.).
36
Although Sextus may be partially dependent on Antiochus in some of his
doxography. See Sedley 2012: 88–93 (and against this Brittain in Sedley 2012: 108–13).
37
DL 7.39–41/LS 26B. Academics used sceptical puzzles to undermine the foun-
dations of ‘logic’; Peripatetics treated it as a tool (organon) of theoretical and practical
 Andronicus of Rhodes 33
order of study had also attracted healthy debate: Zeno and Chrysip-
pus, for instance, had introduced philosophy with logic, Panaetius
and Posidonius with physics. Colourful analogies competed in sup-
port of each position, as philosophy was represented as an egg with
logic for its shell, or a living being with physics for its flesh.
Andronicus would not have accepted the tripartition in such terms,
since he treated logic as a tool serving philosophical inquiry rather
than a proper part of it (see below). Nevertheless, in disputing the
order of study Andronicus and Boethus were both conscious of
contributing new arguments to an old debate. Boethus’ case against
Andronicus builds on Aristotle’s text (ºØ)—a characteristically
Boethan strategy. The quest to understand principles, as Aristotle
suggests, begins from what is familiar to us, and proceeds toward
what is knowable by nature (see for example Physics A.1
184a17–21),38 a course that runs from the particular and perceptible
toward the universal and intelligible (An. Post. A.2, 71b33–72a5). The
particular and perceptible objects of physics (and of the Physics) are
undeniably Æç æ  and ªæØ . So we should begin there.39
Andronicus, on the other hand, seems to have argued for logic’s
pedagogical priority based on the familiar Peripatetic characterization
of logic (cf. DL 5.28) as an instrument or tool (ZæªÆ ):
[W]here to start on Aristotle’s writings . . . Andronicus of Rhodes, the
eleventh of the successive heads of the Aristotelian school, said with
logic . . . Those who said that one should begin with logic asserted that
logic is an instrument (organon), and that one should first know the
instrument and then, on this basis, where the instrument should be
used. For that is how it is in the case of the crafts; the trainee carpenter
first learns the organa, for example the augur and the gimlet, and then

philosophy rather than a proper part (cf. DL 5.28); and Epicureans simply denied that
it had any real function (e.g. Cicero, Ac. 2.97).
38
See Sharples (2010), 43.
39
Andronicus might also have summoned up some textual support for his alter-
native view: Metaphysics Z.4–5, for instance, might be interpreted as suggesting that
logical (logikōs) considerations are more ‘knowable to us’ and are the natural place to
begin philosophical inquiry (1029b7–13). But there is no evidence that Andronicus
relied on such textual authority; cf. Peramatzis 164. The Topics also presents dialectic
as useful to a wide range of disciplines: æd Æ e F æ Ł  æ ºÆ  K
Kø . . . (A.1, 100a18–24).
34 Andronicus on the Categories
on this basis begins on the [craft] of carpentry itself. (Elias, in Cat.
117,17–118,31, tr. Sharples 2010)40
That analogy, though firmly rejected by the Stoics,41 illustrated a
carefully considered Peripatetic position42 that probably developed
in response to Stoic doctrine.43 The analogy highlights a preference
for developing logic just so far as it is useful (åæ Ø : cf. Alex. in An.
Pr. 2,35–4,29) to two real domains of philosophical inquiry, namely
practical and theoretical philosophy, while stopping short of Stoic
hair-splitting and empty formalism.
Like these Peripatetics, and like Antiochus of Ascalon earlier in the
first century (cf. Cic. Ac. 1.30–32, 2.91–92), Andronicus’ interest also

40
  Iæåc H æØ ºØŒH ıªªæÆ ø . . . æØŒ  b ›   Ø  ›
—æØÆ Å ØŒe › ŒÆ  Ø å  B æØ º ı å ºB Ie B º ªØŒB
ºª . . . ƒ b ºª   ‹ Ø E Ie B º ªØŒB ¼æå ŁÆØ çÆ Œ  ‹ Ø ZæªÆ  (20)
 º ªØŒc ŒÆd E æH  NÆØ e ZæªÆ  ŒÆd r Ł’ o ø  F E åæ Æ ŁÆØ fiH
Oæªøfi · o ø ªaæ ŒÆd Kd ÆÆ ø åH åØ· › ªaæ ç Ø Æ N Œ Æ æH 
ÆŁØ a ZæªÆÆ, x  æÆ  æ æ , ŒÆd r Ł’ o ø ¼æå ÆØ ÆP B B
Œ ØŒB (ed. Busse 1900).
41
Generations of Stoics had already rejected the general thrust of this Peripatetic
argument, denying that logic was merely an instrument (e.g. LS 26E): instead, it was
the child of mother philosophy; or the shell of an egg; or the wall of a garden; or the
bones of a living being.
42
The view that logic was a tool of philosophy seems fairly unlikely to have
originated with Andronicus, as Gottschalk (1990: 66 and 1987: 1099) suggests, if the
Stoics considered and refuted the possibility, and especially if the doxography in DL 5
is Hellenistic.
43
Alexander’s treatment in the preface to his commentary on the Prior Analytics
(1,3–2,34) might suggest that the exchange ran along the following lines: Zeno (who
took logic to be a part of philosophy: DL 7.39) resisted an earlier Peripatetic system-
atization that subsumed logic under theoretical philosophy, by objecting that the
subject-matter proper to logic was independent (1,10–2,2); the Peripatetics in their
turn introduced their ‘instrumental’ language (organon not being deployed in this
sense by Aristotle, at least in his school-treatises—unless, as Barnes et al. point out,
Topics 163b9–11 provided them with inspiration) and maintained against the Stoa
that logic was not an independent object of study, because it served a useful purpose
apart from itself, as a hammer facilitates carpentry but is not a part of carpentry; the
Stoa rejoined that philosophy produces logic as a parent produces a child (Ammonius
in An. Pr. 8,20–22, 9,1–2 = LS 26E), whereas tools are not the product of the crafts
which they serve; to which the Peripatetics replied that nothing in fact prevents a tool
from being the product of its own craft, still maintaining the inclusion of logic under
theoretical philosophy (where the doxography in DL 5 still finds it). The course of the
debate—and its reliance on the details of an analogy—may seem a little silly and
tendentious, but it does encode a crucial point: the Peripatetics felt that logical
formalism, pursued too far in its own right (on the perceived line of the Stoics),
could become empty and useless if it was divorced too far from its purpose. See also
Sharples et al. Theophrastus 31.
 Andronicus of Rhodes 35
ran toward the utility of logic (the demonstrative sort of logic that
Antiochus called ‘dialectic’, although Andronicus likely reserved that
title for the practice of the Topics).44 What exactly was logic useful
for? Hellenistic Peripatetics had developed the view that logic served
two functions, namely persuasion and judgement; so runs the (prob-
ably Hellenistic) doxography in Diogenes Laertius:
Logic is developed . . . as a tool (ZæªÆ ). [Aristotle] clearly proposed
that it has two goals, what is persuasive and what is true. He used two
capacities for each of these, dialectic and rhetoric for what is persuasive,
analytic and philosophy for what is true . . . With a view to discovery, he
handed down [to us, in] the Topics and Methodics, a multiplicity of
premises, from which one may have an abundance of persuasive argu-
ments . . . For judgement [he handed down] the Prior and Posterior
Analytics. (DL 5.28)45
Andronicus, as we shall see below, appears to have divided logic along
similar lines (for he argued that a logical treatise by Aristotle, the
Categories, must belong either to dialectic or to demonstration, but
not to both). According to such a view, if you divide logos in two you
will find persuasive speaking, on the one hand, and demonstrative
proof, on the other; a Peripatetically minded student with access to
Aristotle’s school-treatises might choose to study the Topics for the
first, and the Analytics for the second.

44
As an essential ‘judge of truth and falsehood’; similar views can be found (for
example) in Simplicius’ commentary on the Categories. Thus, for example, Androni-
cus celebrated the utility of division (diairesis) as it was practised by the Peripatetics,
and Porphyry followed him (Boethius De Div. 875D): ‘Quam magnos studiosis afferat
fructus scientia dividendi, quamque apud peripateticam disciplinam, semper haec
fuerit in honore notitia, docet et Andronici, diligentissimi senis, de divisione liber
editus . . . ’
45
˚Æd ÆF Æ b ÆP fiH æƪ ı ÆØ ØºÆ.  º ÆØ b K ÆP E ·
Ø e r ÆØ e ŒÆ a çغ çÆ ºª , e b æÆŒ ØŒ, e b ŁøæÅ ØŒ· ŒÆd
F æÆŒ ØŒ F   MŁØŒe ŒÆd  ºØ ØŒ, y   æd ºØ ŒÆd a æd r Œ 
 ªªæçŁÆØ· F b ŁøæÅ ØŒ F   çı ØŒe ŒÆd º ªØŒ, y e º ªØŒe På (5) ‰
‹º ı æ , Iºº’ ‰ ZæªÆ  æ ÅŒæØø . ŒÆd  ı Ø f  Ł 
Œ  f   ØŁÆe ŒÆd e IºÅŁb Ø çÅ .  b æe Œ æ  ı Ø
Kåæ Æ , ØƺŒ ØŒB fi b ŒÆd ÞÅ æØŒB fi æe e ØŁÆ, Iƺı ØŒB fi b ŒÆd çغ çÆ fi
æe e IºÅŁ· Pb  ºØ  h  H æe oæ Ø, h  (10) (29.) H æe
Œæ Ø, h  c H æe åæB Ø. æe b s c oæ Ø    ØŒa ŒÆd Ł ØŒa
ÆæøŒ <ŒÆd> æ  ø ºBŁ , K z æe a æ º Æ Æ ØŁÆH
KØåØæÅ ø x   P æE· æe b c Œæ Ø a ƺı ØŒa æ æÆ ŒÆd o æÆ
(ed. Long 1964).
36 Andronicus on the Categories
1. The Categories introduces demonstration, not dialectic
Bearing in mind this division of persuasive dialectic from rigorous
proof, we can trace a more sophisticated complex of concerns run-
ning behind Andronicus’ organization of the Organon, concerns
which are reflected in Philoponus’ comment that Andronicus began
philosophy from ‘logic, which is concerned with demonstration
(IØØ)’ (in Cat. 5,15–20). Andronicus’ basic concern, I think, is
suggested by his attempt to dissociate the Categories from the Topics
and dialectic, and associate the treatise instead with the Analytics and
demonstration.
During his researches into Aristotelian manuscripts, Andronicus
found something like our text of the Categories prepended to Aris-
totle’s Topics and entitled ‘Preliminary to the Topics’.46 This title and
positioning struck him as a mistake, made by the same troublemaker
who
entitled this book Preliminary to the Topics because he thought these
things [the post-praedicamenta47 beginning at Cat. 10, 11b17 discussing
opposition, priority and posteriority, simultaneity, change, and the
senses of ‘having’] necessary for that work [Top.] in the same way as
the Categories themselves help with understanding the Topics . . . Thus
far Andronicus.48 (ap. Boethium in Cat. 263B, tr. Sharples 2010: 48)
Andronicus reasonably replied that the Topics already has an intro-
duction and does not need another one. More surprisingly, he also
contended that association with the Topics ran ‘against the purpose
(Ææa c æŁ Ø)’49 of the book that he would retitle Predications

46
—æe H ø: Simplicius in Cat. 379,9–11; cf. —æe H ØŒH at 15,30,
16,14 and Porph. in Cat. 57,13.
47
So called because they follow after (post) the central list of ten categories
(praedicamenta).
48
qui hunc libellum Ante Topica <in>scripserit, quod haec ad illud opus necessaria
esse putaverit sicut ipsae Categoriae prosint ad scientiam Topicorum . . . Sed haec
Andronicus (after Shiel 1957 and Moraux 1973: 100 n. 12, correcting Migne 1891).
49
Prothesis, which Porphyry seems to have used in the same sense as Iamblichus’
later prevalent skopos, is an Aristotelian usage: for example, the Topics plainly
proclaims that its prothesis is ‘to discover a method by which we will be able to
syllogize from endoxa about every problem we encounter . . . ’ (A, 100a18–20); in the
Rhetoric Aristotle considers that he will accomplish his prothesis when he has gone
through each of the subjects he promised to discuss (957a30); and the Categories itself
uses the word in a somewhat similar sense (‘someone might say’ that Aristotle has
gotten off the beaten track by discussing relatives when ‘we proposed to discuss
quality [ bæ  Ø Å  c æŁ Ø  ØÅ Æ ı, 11a20–1]’).
 Andronicus of Rhodes 37
or Categories (Simplic. in Cat. 379,9–10), perhaps in order to empha-
size its role as a training-text in predication, itself a propaedeutic to
division and demonstration. His latter remark invites scrutiny: while
it is easy to grant to Andronicus that the Categories was not written
for the purpose of introducing the Topics, it is odd to conclude that
the Topics is irrelevant or somehow distracting from the purpose of
the Categories. The Categories, after all, could be read as an elemen-
tary handbook for an important component of the practice that
Aristotle calls ‘dialectic’ in the Topics: it offers intuitive tests for
establishing under which genus any term falls.50
A parallel passage in Porphyry’s shorter surviving commentary on
the Categories, I think, expresses what Andronicus had in mind
(56,22–31):
It would be absurd to call the book Introduction to the Topics, for why
call it Introduction to the Topics rather than Introduction to the Analy-
tics or Introduction to On Interpretation? It is not for the sake of
studying the Topics that one first has to learn about predications
(ŒÆ Ū æÆØ), but also for the sake of learning about the Analytics and
about categorical propositions and just about any other subject. This
work is the most elementary one, and serves as an introduction to all the
parts of philosophy.51
Some later Peripatetics rejected this view and defended the title —æe
H ØŒH.52 But this passage and its counterparts in Simplicius
and Boethius seem like (a precis of ) the only serious argument
preserved against that title.53 Andronicus appears to be the strongest

50
On the association, see Menn 1995, Bodéüs 2001: lxiv–lxxxix.
51
—æe b H ØŒH I ø ¼ Ø Kتæç Ø· Øa  ªaæ —æe H ØŒH,
Iºº’ Påd Aºº  —æe H Iƺı ØŒH ŒÆd —æe F æd æÅÆ; P ªaæ Øa c H
ØŒH ØÆ ŒÆºÆ æ ÆŁØ E a H ŒÆ Ū æØH, Iººa ŒÆd Øa c H
Iƺı ØŒH ŒÆd Øa c H ŒÆ Ū æØŒH æ  ø ŁÅ Ø ŒÆd åe Øa a ¼ººÆ
 Æ ÆŁ Ø· ØåØø Æ  ªaæ F ŒÆd N ƪøªØŒe N  Æ a æÅ B
çغ çÆ e غ . ŒÆd ºØ Æ æe F çı ØŒ F i YÅ æ ı B çغ çÆ j
æe H ØŒH· ç ø ªaæ æª  P Æ,  Øe ŒÆd a ‹ ØÆ. —æd b H ªH F
Z  j —æd H ŒÆ ªH PÆH åæc KتæçØ (ed. Bodéüs 2008).
52
See Simplicius’ discussion at in Cat. 15,26–16,5, and Elias in Cat. 241,30.
53
On Porphyry’s view, the preliminary study of predications is useful both for the
Analytics and the Topics. Andronicus would presumably have allowed this as well: it
would have been very odd for him to insist that the study of predications was useful
only for the Analytics. But he clearly rejected an exclusivist position that the study of
predications was useful only for the Topics. (It is not clear that anyone ever held such a
position, but Andronicus took some such argument to underwrite the title Before the
Topics.) For Andronicus’ role in the title controversy, see Bodéus (2001: xxv). The
38 Andronicus on the Categories
candidate for the origin of the argument,54 although Porphyry does
not name him here.55 The point is simply that the Categories is not
(only or primarily) an introduction to the Topics’ dialectic, the art of
reasoning from plausible endoxa without self-contradiction. Instead,
its primary function is to introduce demonstrative argument—that is,
the Analytics. It is not yet clear why this should be so, but it does seem
clear that Andronicus wishes to associate the Categories with dem-
onstration and to dissociate it from dialectic, or anything else ‘against
its purpose’ (Simplic. in Cat. 379,9–10), perhaps including rhetorical
uses more broadly. (Indeed, the list of ten categories may well have
been in rhetorical use already.)56
Furthermore, there seems to be a good deal at stake for Andronicus
pedagogically in this association, since he thinks that philosophy
begins with logic precisely because logic includes an education in
demonstration (Philop. in Cat. 5,19), and that training in this ‘tool’
will enable us to apply judgement and discover the truth in theoretical
and practical philosophy (for this view, cf. Simplicius in Cat. 6,4–5,

argument is matched, with trivial variations, at Simplic. in Cat. 15,26–35 (and,

renewed in response to Adrastus, from 15,36–16,16), and Boethius in Cat. 162C–D,
263B, etc.
54
Andronicus, one might suggest, must have offered an argument; if he offered the
only argument in the tradition, this is it; if someone else offered an importantly
different argument, we could reasonably expect to find them both preserved, but this
is all we find.
55
Porphyry here mentions On Interpretation, which Andronicus athetized; but the
substance of the argument has nothing to do with De Int., focusing instead on the
function of Cat. as an introduction to the Analytics and demonstration, designedly
contrasted against the Topics and dialectic. It is also worth noting that De Int. is
mentioned after the Analytics here and in the parallel passages; if Andronicus had
argued that the Cat. constituted a good argument to the Analytics and demonstration,
a later commentator steeped in the tradition that embraced De Int. between Cat. and
An. might easily have added ‘ . . . or Introduction to On Interpretation’ to the skeleton
of Andronicus’ case against the title —æe H ø. (It might also be added that
Andronicus only marked the De Int.’s authorship as questionable: he would not have
deleted it from his catalogue, and he need not have deleted it from his logical
curriculum; he was, after all, prepared to introduce other non-Aristotelian innov-
ations.) It is perhaps interesting to note that the De Int. itself has been read as a work
primarily about dialectic and the study of contradictory premises; if Andronicus
interpreted it along similar lines, he might have supposed it had no place between
Categories and Analytics, both being concerned with demonstration and not with
dialectic.
56
Quintilian, at least, bears witness that the list of categories were in rhetorical use
close to Andronicus’ lifetime: Inst. 3.23–34. cf. Bodéüs 2001: xxii.
 Andronicus of Rhodes 39
Ammonius 6,3–8, Olympiodorus 9,9–11), whereas dialectic will be
less useful for this function. Thus Andronicus, like many later stu-
dents of Aristotle, strives to highlight a method ‘stronger’ than dia-
lectic in Aristotle.57
Can we obtain any insight into the reasons behind Andronicus’
association of the Categories with demonstration? First, to set the
scene, we should notice that the endorsement of logical demonstra-
tion as a way to the truth, emphasizing division and definition, was a
long-lasting Stoic commonplace. The Stoics had championed the
value of dialectic through the Hellenistic period, including ‘definition
and division and neat syllogistic proof ’ (e.g. Cicero Ac. 1.5), in order
to get at the truth of things. Antiochus endorsed this Stoic method-
ology and wrapped it into the collective practice that he attributes to
the ‘Old Academy’, embracing Plato and Aristotle as well as the Stoics
themselves (Cicero, Ac. 1.30–2; cf. Karamanolis 2006, 62–71): Antio-
chus stressed how the followers of Plato pursued definitions, and
etymology, in order to obtain more certain knowledge (scientia)
than was unavailable in the transient realm of particular things.58
(Cicero, speaking on behalf of Philonian scepticism, believes that
sophistical puzzles turning on ambiguity represent a serious challenge
to the Antiochean celebration of dialectic. Similar puzzles resurface
later in criticism of the Categories.)59

57
Compare for example Irwin 1988, who distinguishes between ‘weak’ and ‘strong’
dialectic.
58
Cicero introduces the notion of species (idea) and explains that the ‘Old
Academics’—including Platonists and Aristotelians—‘approved the method of defin-
ing things’ (definitiones rerum) in order to gain knowledge (scientia) unavailable from
the transient world of particulars, and thus ‘approved the explication of words
(verborum explicatio), the statement of the reason why each class of things bears the
name that it does—the subject termed by them etymologia . . . and under this head
was imparted their whole discipline of Dialectic’ (tr. Rackham/van den Berg 2008).
59
Cicero offers the following reply to this Antiochean celebration of dialectic. The
pro-dialectic crowd, he suggests, has not yet managed to resolve sophistical puzzles
turning on ambiguity—such as the sophistical syllogism proposing that a coat is
manly, but being manly is brave, so a coat is brave—nor puzzles turning on equivo-
cation, nor old chestnuts like the liar or sorites, nor other cases of ambiguity, in
response to which Chrysippus had famously advocated a policy of ‘quiescence’
( ıåÇØ): Ac. 2.92–6 = LS 37H; cf. Atherton 1993, 180–3, and Bobzien 2005 for
the Stoic response to puzzles of this kind. Note that Nicostratus cites such a puzzle
against the Categories in the newly discovered fragmentary commentary
(Chiaradonna, Rashed, Sedley, and Tchernetska 2013, 1,26–3,1).
40 Andronicus on the Categories
Andronicus will inhabit this Antiochean atmosphere,60 exhibiting
confidence in the value of dialectic and division, and demonstration,
and tracing these methods to the ancient authority of the Old Acad-
emy. Andronicus too is sympathetic to the methods of dialectic in
this Stoic sense,61 and particularly divisio or ØÆæ Ø, as we have
seen both in the Greek sources and will see below in Boethius
(De Divisione). Andronicus is also explicitly sympathetic to some
views adopted by the Old Academic Xenocrates (cf. Simplic. in Cat.
63,22–8, Themistius in De An. 31,1–32,34). In the generations fol-
lowing Andronicus, however, there is something new, not (so far as
our evidence suggests) to be found in Antiochus: where the virtues of
division and demonstration are extolled, the ‘ten categories’ are now
described a crucial tool, whose value was anticipated by both Plato
and Aristotle. Plutarch (perhaps drawing here on Eudorus of Alex-
andria) and Alcinous offer illustrative examples of this development:
Again he [Plato] gives indications of the ten categories both in the
Parmenides and elsewhere, and in the Cratylus he goes thoroughly
into the whole topic of etymology. In general, the man was supremely
competent in, and a connoisseur of, the procedures of definition,
division <and analysis>, all of which demonstrate particularly well the
power of dialectic. [ . . . ] The imposition of names is according to the
nature (ç Ø) of the reality (æAªÆ). [ . . . ] The name is an instrument
which teaches about and divides the essence of each reality, as the
shuttle does for the weaving of cloth [cf. Cratylus 388B13–C1].62 (Alci-
nous, Didasc. 6, 159,43–160,30, tr. Dillon, 1993).
Not only do the souls of mortal beings possess the capacity to know
the sensible (ªø ØŒc F ÆN ŁÅ F ÆØ), but [Plato] adds that the

60
For Antiochus and his ‘Platonism’, see Chiaradonna 2013, Bonazzi 2012, and for
an overview of his philosophy in general terms Sedley 2012. For Antiochus’ possible
influence on Andronicus, see Donini 1982: 83–4 and 90–1 on Andronicus’ ‘Platon-
ism’; for general similarities between Andronicus, Boethus, and Antiochus, see
Karamanolis 2006: 82.
61
On Stoic dialectic, see LS 31; for the Stoics’ treatment of earlier ‘dialectic’, see
Brunschwig 1991.
62
˚Æd c a ŒÆ ŒÆ Ū æÆ   fiH —ÆæfiÅ ŒÆd K ¼ºº Ø Ø, e
K ı º ªØŒ    ‹º  K fiH ˚æÆ ºø fi Øæå ÆØ· ±ºH b ƒŒÆ Æ  › Icæ
ŒÆd ŁÆıÆ c B  ›æØ ØŒB ŒÆd ØÆØæ ØŒB æƪÆ Æ, ÆQ A ÆØ Œı ÆØ
ºØ Æ c ÆØ B ØƺŒ ØŒB [ . . . ] u  r ÆØ Æ e Z Æ ŒÆ a e NŒE 
fi B F æªÆ  ç Ø Œ · [ . . . ] ! 0E Ø ªaæ e Z Æ ZæªÆ  æªÆ  På n
 ıå, Iººa e ŒÆ ººÅº  fi B ç Ø· ŒÆd Øa  ı Ø Œ  Iºº º ı a
æªÆ Æ ŒÆd ØÆŒæ  ÆP , u  r ÆØ e Z Æ ØÆ ŒÆºØŒ Ø ŒÆd ØÆŒæØ ØŒe
B Œ ı P Æ ZæªÆ , ‰  ŒæŒd F ç Æ  (ed. Whittaker 1990).
 Andronicus of Rhodes 41
soul of the cosmos—whenever she touches the scattered being or the
undivided being of anything—is moved throughout her entire being
and announces what the object is identical with, and from what it is
different, and in what relation (æe ), and where and how, and when,
it comes about that each thing exists (r ÆØ) and is acted upon by others
( åØ), both in the sphere of becoming and in that of the ever-one. In
these words he is also giving an outline of the ten categories; likewise in
what follows, he makes the case more clearly . . . 63 (Plutarch, de proc.
an. 1023D–E)64
There is at least a prima facie case that Andronicus may have played a
role in this development, absent in Antiochus but growingly import-
ant in the following generations, by ‘foregrounding’ the Categories in
his catalogue of Aristotle’s technical treatises, representing it as a
crucial tool for demonstration and division. I think that this case
can be strengthened and clarified by exploring the role that division
plays in Andronicus’ view of demonstration.

2. The Categories introduces demonstration by

cultivating the use of division to
construct essential definitions
This section falls into two parts. (a) In the first part, I offer a more
general exploration of the fragmentary sources for Andronicus’ inter-
est in the value of division, focusing on the importance of distin-
guishing per se from per accidens predication, and drawing primarily
on Boethius’ treatise De Divisione.65 Evidence for the relationship of

63
ŒÆd c P   ƃ H ŁÅ H łıåÆd ªø ØŒc F ÆN ŁÅ F [E] ÆØ
å ı Ø, Iººa ŒÆd c F Œ  ı çÅ d [Tim. 37A] IÆŒıŒº ıÅ ÆP c æe
Æı , ‹ Æ P Æ ŒÆ c å   Ø  Kç Å ÆØ ŒÆd ‹ Æ IæØ , ºªØ
ŒØ ıÅ Øa  Å Æı B, ‹ fi ø ’ ¼ Ø ÆP efi q ŒÆd ‹ ı i " æ , æe ‹ Ø 
ºØ Æ ŒÆd ‹Å ŒÆd ‹ø ıÆØ ŒÆ a a ªØªÆ æe "ŒÆ  <"ŒÆ Æ> r ÆØ
ŒÆd  åØ. K  Ø –Æ ŒÆd H ŒÆ ŒÆ Ū æØH  Ø    ªæÆçc  Ø
Aºº  E KçB ØÆ ÆçE (ed. Hubert 1954).
64
This text mentions Eudorus several times previously, in connection with the
transmission of Academic views on the Timaeus and the construction of the soul, and
quotes or cites Eudorus at several points in the treatise; as we know from other sources
that Eudorus was interested in the Categories, expounded his interest Pythagorically
(based on the tenfold nature of reality), and influenced or was influenced by Andro-
nicus of Rhodes, this attribution of the ten categories to Plato here might also be
ascribed to Eudorus. For Eudorus’ possible influence here, see also Bonazzi 2013.
65
On the evidence in Boethius for a monograph by Andronicus On Division, see
the excellent introduction to Magee’s 1998 edition, with Moraux 1973: 120 ff. Donini
42 Andronicus on the Categories
these views to the Categories is circumstantial, but bolstered, I believe,
in the following section. (b) In the second part, I examine the sources
specifically for Andronicus’ overarching division of the ten categories
into two, absolute and relative (ŒÆŁ! Æ  and æe ), and suggest
that the motivations for this division might be understood in terms of
the sources offered in part (a); in this discussion I rely primarily on
the evidence of Simplicius. Broadly, I argue that, on Andronicus’
view, the Categories helps us to distinguish between essential and
accidental kinds of predication without prior training in logic, in
order that we can begin to study division, and eventually construct
good definitions based on essential differentiae rather than accidental
features. The Categories’ fourfold division (1a20–1b9) and sketch of
the attributes that characterize substances ( P ÆØ, 2a12–4b19) in
contrast to non-substantial predicates (4b20–11a39) are understood
as a framework for the application of Aristotelian essentialism in
dialectic or demonstration.

(a) Andronicus on division

Good ancient evidence strongly suggests that Andronicus saw div-
ision as crucial for the construction of good definitions, and so for
demonstration (cf. Donini 1982: 90–2, Gottschalk 1987: 1115). Com-
bined testimony from Boethius, Porphyry, and Simplicius offers
helpful guidance for deducing Andronicus’ position. Recall, Boethius
informs us that Andronicus especially praised the art of division
(scientia dividendi, ØÆæ Ø):
The book On Division published by Andronicus, a most diligent scholar
of old (diligentissimi), treats of the considerable advantages the science
of dividing brings to scholars and of the high esteem in which this
branch of knowledge was always held within the Peripatetic school.
Plotinus, a most profound philosopher, thought highly of Andronicus’
book and Porphyry adapted it in his commentary on Plato’s dialogue
entitled The Sophist. It was also Porphyry who acknowledged the utility
of his Introduction to the Categories with reference to this science.

1982: 90–2 and Gottschalk 1987: 1115 argue that Andronicus intended to connect
division and definition. Mansfeld 1992: ch. 5 offers a very helpful excursus on
substance, division, and the Categories, including the early commentary tradition:
see 74–5 on Andronicus’ De Divisione.
 Andronicus of Rhodes 43
For he says that a knowledge of genus, species, difference, property, and
accident is a necessary prerequisite to, among other things, partitioning,
which is of the greatest utility.66 (De Div. 875D–6D, lightly modified
from 5,1–11 Magee, 1998)
Andronicus’ evaluation of division as a useful technique for philoso-
phy67 may be partially motivated by the appreciation that he
expresses for demonstration (IØØ). As we have seen above,
Andronicus believed that mastering Aristotle’s Analytics would be
fundamental to a grasp of scientific demonstration, and as a good
reader of the Analytics, he would certainly have come away with the
impression that the art of division and demonstration are closely
intertwined (see for example An. Post. 2.5, 2.13), even if division
cannot alone achieve all the results that the Academics had claimed
for it (cf. An. Pri. 1.31). Andronicus’ appreciation for ØÆæ Ø also
proved highly influential into late antiquity:68 his influence on Plo-
tinus and later Neoplatonists, especially Porphyry, is also attested in
other contexts (cf. Wilberding 2005: 452). I argue here that this
emphasis on division will help to understand Andronicus’ reasons
for treating the Categories as a crucial introduction to the Analytics.
Of particular relevance for us here, in a passage from the same
treatise On Division that John Magee (1998) has persuasively argued
to reflect Andronicus’ views, Boethius writes:
The later sect of Peripatetic wisdom [sc. Andronicus]69 discerned in the
most diligent manner (diligentissima) the difference between divisions:

66
Quam magnos studiosis afferat fructus scientia diuidendi quamque apud Peri-
pateticam disciplinam semper haec fuerit in honore notitia, docet et Andronici
diligentissimi senis De diuisione liber editus; et hic idem a Plotino grauissimo
philosopho comprobatus et in Platonis libri qui Sophistes inscribitur commentariis
a Porphyrio repetitus, et ab eodem per hanc Introductionis laudata in Categorias
utilitas. (ed. Magee 1998).
67
See also Donini 1982: 90–1.
68
Marwan Rashed has also drawn attention to the influence of Andronicus’ views
on ØÆæ Ø, as preserved in Boethius, on his ontology, as it was criticized by
Alexander of Aphrodisias in two unedited fragments surviving in Arabic (Rashed
2004). Porphyry’s appreciation for ‘division’ in the Isagoge, where he follows various
‘old masters’ (1.8, 1.15), might follow the same source as his discussion of division in
his Sophist commentary alluded to by Boethius.
69
On the association with Andronicus, see Magee 1998 ad loc. (167–8) and his
introduction. The evidence is circumstantial, but strong. Boethius, reporting a trad-
ition that presumably stems from Porphyry, believes ‘that the later sect and Andro-
nicus form a unity . . . although the historical accuracy of his view is in the end a
question of the reliability of his source, Porphyry, or of his interpretation thereof ’
44 Andronicus on the Categories
it separated division per se and division secundum accidens from one
another, and distributed them both. Its predecessors, on the other hand,
indiscriminately employed both an accident in place of the genus
and accidents in place of species, or differentiae.70 (De Div. 891–2,
48,26–50,5 Magee)
The closing sentence of this passage suggests a criticism of earlier
Peripatetics’ lack of care in distinguishing between per se and per
accidens predication, which in some way vitiated their practice of
division. By contrast, the later Peripatetics methodically organized the
differentiae used in divisions into the correct silos, recognizing which
apply per se and which apply per accidens, and further subdividing
these two classes (I take this to be the force of distribuit at 48,28–50,1,
perhaps generating the kind of ‘tree’ offered by Boethius himself in
the De Divisione).71
Some illustration will be useful here. Suppose that I am investigat-
ing what a human being (¼Łæø ) really is (see for example Meta-
physics 7.17, 1041a25–32), that I have arrived at the genus ‘animal’,
and that I plan to divide this genus using differentiae in order to
arrive at a specific definition (cf. Topics 6.4, 141b26). Suppose next
that I use differentiae such as the following: sleeping, sitting, standing,
being awake, curly-haired, and grey-eyed.72 I will not get far in
Aristotelian science this way, for these accidental or incidental
differences—attributes that an animal might lose or gain without
ceasing to exist as a member of its kind—are unsuitable for arriving

(Magee 1998: 167). Diligentissimus described Andronicus at 4,5 Magee, ‘which sug-
gests that Andronicus is the figure intended here as well’ by the descriptive diligen-
tissima at 48,27. Other sources suggest that the division of ‘later’ Peripatetics
(posterior . . . Peripateticae secta) belongs to the first-century bce (based on Strabo
Geog. 13.1,54 and Aspasius in Eth. Nic. 44,20). I would add the passage Simplic. in
Cat. 63,22–6, quoted below, where Andronicus is associated with a division of the
Categories into absolute and relative, or substantial and accidental; I will discuss this
further below.
70
Posterior quidem Peripateticae secta prudentiae differentias diuisionum diligen-
tissima ratione perspexit et per se diuisionem ab ea quae est secundum accidens
ipsasque inter se disiunxit atque distribuit, antiquiores autem indifferenter et accidente
pro genere et accidentibus pro speciebus aut differentiis utebantur . . . (ed. Magee 1998).
71
Diuisio secundum se may be genus in species, totum in partes, uox in significa-
tiones; diuisio secundum accidens may be subiectum in accidenta, accidens in subiecta,
accidens in accidentia (6,17–10,27 Magee).
72
Such is the list of per accidens differentiae at Boethius De Div. 18,4–8 Magee,
combining a list of regularly departing (statim relinquentes) and consequent (con-
sequentes) accidental differences.
 Andronicus of Rhodes 45
at definitions (see for example Ar. An. Post. 2.5 and 2.13, with
commentary in Barnes 1993; Ar. Parts of Animals 1.2–3; Ar. Topics
1.5, 101b38, Boethius De Div. 880D–1D, 16,16–20,19 Magee).
Instead, when I undertake to define a reality or substance ( P Æ),
I should be careful to predicate only in the first of the ten categories,
 K Ø (‘what it is’; note that Andronicus may not have distinguished
carefully between the Categories’ and Topics’ accounts of the first
category, as briefly noted below). Consider Aristotle’s exploration of
how we might exhibit the essence of a subject demonstratively using
division (An. Post. 2.13, 96a20–97b2): to predicate only in the first
category, as Aristotle puts it, requires that I ‘establish things through
the genus’ (Øa F ª ı ŒÆ Æ Œı ÆØ, 97a27–8), discarding inci-
dental features in the other nine, non-substantial categories like
‘musical’, ‘pale’ (73b5), and the colourful examples cited above by
Boethius, and focusing just on what the thing is ‘in itself ’ or per se
(ŒÆŁ’ Æ : cf. An. Post. 1.4, 73a34–b5). (When Andronicus speaks of
per se predication, he seems to have in mind predication in the first
category, which he also treats as a category of substance ( P Æ: see
also Simplic. in Cat. 63,22–6, discussed below). Thus Andronicus
mostly has in mind examples like ‘the human being is an animal’
when he speaks of ‘per se predication’, while he has in mind examples
like ‘the human being is pale’ when he speaks of ‘accidental predica-
tion’. Andronicus, perhaps like Aristotle in his later development,
might restrict the category of ‘what-it-is’ to substances.)73
On this interpretation, when I focus just on essential predicates
belonging in the first category, focusing my investigation on what
belongs to the substantial subject per se, these predicates will prove to
represent features that are basically explanatory of my subject and
possessed by it in the special sense that, if the subject were to lose
those features, the subject would cease to exist. If I investigate the
question carefully as it relates to the example of ‘human’ above, I will
find that rationality is such a fundamental or essential feature of
humanity, a sine qua non for being a human being. That is to say, it
holds of a human being ‘in itself ’, and I would be able to narrow down
a scientific definition of humanity from genus and difference: for
example, ‘rational animal’.

73
See Frede 1987: 45, and more broadly, 29–48, on the development of the
‘doctrine of categories’.
46 Andronicus on the Categories
There are various ways in which Aristotle thinks that we could go
awry in executing this kind of project and so fail to lay the correct
groundwork for a good scientific definition, but the first and foremost
error is to fail to distinguish those terms that belong to the first
category,  K Ø, and therefore can be predicated of my subject per
se, from those terms that refer to accidents or incidental features. This
seems to be the kind of failure with which Andronicus charges
‘earlier’ Peripatetics: possibly he criticized them much as later Aris-
totelian commentators would criticize the Stoic practice of division
and definition, as taking ‘the definition to be no different from the
representation of the peculiar characteristic’ or proprium (Alex. in
Top. 42,27–43,2: see LS 32E), and thus allegedly losing track of
essentialism altogether. A counter-offensive in favour of Aristotelian
essentialism, at least in the practical task of developing definitions,
might have begun with Andronicus, whose treatment of at least some
Aristotelian categories—such as the relative, discussed further
below—shows evidence of possible Stoic influence.74
Be that as it may, we are primarily interested in the role of the
Categories in Andronicus’ view. The core of the treatise that Andro-
nicus came to call Categories (the so-called praedicamenta) offers a
series of heuristic recipes for just this kind of practice, namely,
determining which terms fall under which genera and so enabling
the construction of essential definitions. (Indeed, the Categories may
be the only place in the corpus where one can go for a sustained
account of such a series of recipes for the categories.)75Andronicus
would have found here a wealth of relatively non-technical outline
accounts (  ªæÆçÆ)76 of the genera for getting this right. For

74
It might seem surprising for Andronicus to accuse Peripatetics of ignoring such
a fundamental Aristotelian position as essentialism, and I would not want to push this
point too far. But (a) it may well be that he regarded his predecessors as endorsing
something like Stoic logic over Aristotle’s (and, as Alexander would also stress, the
Stoics fail to distinguish between essential and accidental features), and (b) if Andro-
nicus himself had anything to do with promoting the story that his predecessors had
disregarded or underemphasized the ‘esoteric’ treatises that he was now publishing, it
would stand to reason that Andronicus might also accuse them of ignoring important
doctrines in those treatises.
75
For this function of the Categories, once again, see Bodéüs 2001: xxxix and Menn
1995: 319–20.
76
As Jonathan Barnes has suggested to me in correspondence, a  ªæÆç of Fs is
a proposition of the form ‘Fx iff Gx’ which is informative and true, but not a
definition. I argue below that Andronicus took such  ªæÆçÆ as more accessible
to the beginner than a definition, easily explained in terms of everyday experience.
 Andronicus of Rhodes 47
instance, I can test whether the item referred to by a given term is a
secondary substance or a non-substantial feature by checking
whether it is IN the subject, in the technical sense that it is in it ‘not
as a part, and unable to exist apart from what it is in’ (Cat. 1a24–25):
if the answer is affirmative, we have here a non-substantial attribute,
such as a quality (say, the paleness of Socrates’ body, or the universal
quality colour to which that paleness belongs), which will not help us
to construct a scientific definition. If we were not equipped to make
these clear distinctions—that is, if we were unable consistently to
locate the category to which a term belonged—the central demon-
strative project of definition would become impossible; and in this
way, having lost our capacity to construct definitions, we would be
prone to slip up in many ways.
The Categories has often appeared to be useful for making it
possible to speak about the essential features of ordinary, perceptible
things (see Mann 2000: 184–204). In general, someone like Andro-
nicus who wished to emphasize the distinction between essential and
accidental predication would find particular value in promoting Cat-
egories 2 (1a20–1b9)’s celebrated fourfold division of items SAID-OF
a subject and IN a subject (the following table is adapted from
Reinhardt 2007: 515, also discussed above in the Introduction):

Cat. 2, 1a20–b9 Beings IN a subject Beings not IN a subject

Beings SAID-OF a Genera and species of Genera and species of primary

subject non-substantial items
Beings not SAID-OF Individual non-substantial
a subject items, such as  Ø
ªæÆÆ ØŒ (‘some particular
knowledge of grammar’)
substances (i.e. secondary
substances)
Primary substances, individuals,
e.g. › Ø ¼Łæø  (‘some
particular human being’)

In the project of generating divisions that construct good definitions,

in turn facilitating demonstration, the Categories would appear useful
on several fronts. The reader may learn to narrow down what counts

I stress the word  ªæÆçÆ here because I will argue below that this ‘non-technical’
nature of the descriptions of the genera in the Categories made it especially useful for
Andronicus’ introductory purposes.
48 Andronicus on the Categories
in the first category, building from the Categories’ clear, pre-technical
outline accounts or  ªæÆçÆ for recognizing the genus under which
a given term falls, especially the fundamental fourfold division here in
Cat. 2 and the account of P Æ in ch. 5. Moreover, the cautionary
division of homonymy from synonymy and paronymy in ch. 1 would
help the reader to resist sophistical puzzles based on ambiguity and
equivocation—the kinds of puzzles which, as we noticed above, may
have troubled Antiochean advocates of dialectic and demonstration.

(b) Andronicus on the reduction of ten categories

to absolute and relative
Moreover, Simplicius tells us that on Andronicus’ view, the ten
categories described in the central section of the Categories were
embraced by two broader headings, absolute (ŒÆŁ! Æ ) and relative
(æ Ø) (perhaps adopting a view previously adopted by Xeno-
crates77 based on the fundamental division of absolute from relative
in Plato Sophist 255D). Simplicius explains that:
Xenocrates and Andronicus and their followers seem to include all [the
ten categories] in [the opposition] ‘by itself ’ (ŒÆŁ! Æ ) and ‘relative’
(æ Ø) . . . 78 [they say] that accidents are ‘relative’ as [being] always of
other things, and that substance is by itself.79 (Simplic. in Cat. 63,22–6)

77
On the relation between Andronicus and Xenocrates, see also Tarán 1981: 741–2
(commenting on Moraux’s view). For a reconstruction of Xenocrates’ own position,
see Dillon 2003: 150–1.
78
Simplicius suggests that Andronicus and Xenocrates thought the division into
ten was excessive or ‘superfluous’, but Moraux and others have shown, I think
persuasively, that Andronicus did not seek to replace the ten categories with two, as
he also maintained the validity of the tenfold division. Rather, he seems to have sought
to understand the tenfold division in terms of a twofold division. After all, Aristotle
himself at Metaph. ˜ 5 (1017a23–30, below) makes a first division of being into the
kata sumbebēkos and kath’ hauto, before classifying the skhēmata tēs katēgorias under
the kath’ hauto; once the kata sumbebēkos is identified with the pros ti, it seems simple
to find the subordination of the ten to the two in Aristotle himself. If Andronicus
maintained that the Categories and the Metaphysics express one and the same
doctrine, he presumably would not have regarded such a subordination as a betrayal
of the project of the Categories.
79
ƒ ªaæ æd ˛ Œæ Å ŒÆd æØŒ   Æ fiH ŒÆŁ’ Æ e ŒÆd fiH æ Ø
æغÆØ  Œ F Ø, u  æØ e r ÆØ ŒÆ ’ ÆP f e F  H ªH
ºBŁ . ¼ºº Ø b N P Æ ŒÆd ıÅŒe ØÆ  ı Ø· ŒÆd y Ø b ÆP  ø
 Œ F Ø E æ æ Ø ºªØ a ıÅŒ Æ æ Ø ºª ı Ø, ‰ ¼ººø Id Z ø,
ŒÆd c P Æ ŒÆŁ’ Æ  (ed. Kalbfleisch 1907).
 Andronicus of Rhodes 49
Andronicus, then, subsumes the ten ŒÆ Ū æÆØ under these two
broad classes as the ‘highest division’ in the Categories, a move that
his successors would endorse and follow.80 Moreover, Simplicius goes
on to charge Andronicus and Xenocrates with reducing all of the
genera of accidents to the relative (in Cat. 63,28–30), suggesting that
in some way Andronicus lumped quantities, qualities, and the rest
together as relatives. The passages from Boethius discussed in part (a)
above might help us to make sense of this view: Andronicus wishes to
distinguish carefully between items predicated of a subject essentially,
in the first category (  K Ø) (which Andronicus may restrict to
substances), and items predicated of a subject accidentally, in the
nine non-substance categories. Naturally (given that he believes that
Categories mirrors An. Post. and that  K Ø is the first category of
both), he finds the Categories to be a useful place to look for elabor-
ating this distinction.
But we have conflicting evidence about the exact position adopted
by Andronicus concerning the status of the category of the relative
(æ Ø), and there has been considerable debate over where he
stands.81 On the one hand, as we have seen, Andronicus sought to
reduce the number of categories to two, absolute and relative. On this
view, if ‘relative’ is taken as amounting to ‘accidental’ (see above,
63,22–6), there would be just two ways in which one might predicate
a feature of Socrates: either in the first category (for instance, ‘Socrates
is human’) or accidentally (‘Socrates is wise’ or ‘pale’). Then again,
Andronicus also aimed to preserve the number of ten categories
(Simplic. 342,21–5), so that he did not want to jettison quantity,

80
While this has the historically important effect of integrating the ‘Platonic’
[Sophist 255D] and ‘Aristotelian’ categories, it also has some Aristotelian support:
see Metaph. ˜7, 1017a23–30, and Bodéüs 2001: lxxxi–ii. Andronicus’ move was very
influential: we have later evidence that the ‘Andronican’ division prevailed in the
subsequent tradition as the ‘highest cut’ in the Categories, perhaps with substance
functioning as absolute and the accidents as relative. Pseudo-Archytas, who may have
written in the first century bce as a near contemporary of Andronicus and follower of
Eudorus (cf. Szlezák 1972), divides being into subjects and sumbebēkota just before
rostering the ten categories. Alexander also called this the ‘highest division’ in the
Categories ( fi B æ fi Å ØÆØæ Ø H K ˚Æ Åª æÆØ, in Met. 242,15–16 and 243,3), as
does Porphyry (in Cat. 71,28). Indeed, Alexander, Porphyry, and the subsequent
Neoplatonic ‘consensus’ seem to follow Andronicus’ basic view about the Categories
here, as Boethius suggests, and Andronicus may well be one of the ‘older masters’
followed in Porphyry’s Isagoge (1,3–17).
81
See Mansfeld 1992: 59–61 for a helpful summary, and Reinhardt 2007 for an
especially clear and succinct presentation of the issues.
50 Andronicus on the Categories
quality, relation, and the rest as valid kinds of predicates. And yet
again, he regarded the relative as a kind of ‘offshoot’ that belonged
after the other categories (157,18–22; cf. Aristotle, Nic. Eth. 1.6,
1096a21), so that it seems on the face of it unlikely that he simply
meant (say) to subordinate the remaining eight non-substance cat-
egories under the relative, so that he could have his cake (maintaining
quantity, quality, relation, and so forth) and eat it too (subsume these
under a single umbrella, relation).
Also relevant here is Andronicus’ proposal (shared with his con-
temporary Ariston) to amend Aristotle’s second definition of the
category of the relative (æ Ø) at Cat. 8a31, in order to require
that the item described exists only insofar as it relates to something
different from itself.82 The ostensible purpose for this revision is to
avoid vicious circularity, which could be a reasonable motive in its
own right; but the correction also brings Aristotle’s (second) defin-
ition of the relative into concinnity with the fourth ‘category’ devel-
oped by the Stoics, the relatively disposed (æ Ø ø å ). This has
led some scholars (such as Tarán 1981: 741–2) to suggest that An-
dronicus is directly influenced by the Stoics.83
There are several ways in which we might make sense of the
differing evidence for Andronicus’ view.84 I propose the following

82
From items ‘for which being is the same as being somehow related to something’
(æ Ø) to items ‘for which being is the same as being somehow related to something
else (æe " æ ) (see Simplic. in Cat. 201,34, Porph. in Cat. 125,14–23).
83
On this ‘category’ and for what follows, see Menn 1999. This kind of property is
weaker and, in a sense, further removed from real being than the other Stoic
‘categories’, such as the  Ø, in the following sense: some attribute F (say, the
attribute of being someone’s sibling) is a æ Ø ø å  just in case any item that
is F has no intrinsic difference from an item that is not F. Such attributes are such that
‘they can belong to something and then not belong to it without any change or
alteration in the thing’ (Simplic. in Cat. 166,17–19, discussed by Menn 1999: 232),
like a case of what we might now call Cambridge change. But, Richard Sorabji points
out to me in conversation, David Sedley has shown (2000) that the Academy had
already introduced the concept of the æ Ø ø å  deployed by Aristotle at Cat.
8a31–2; Andronicus may be reflecting earlier Academic views here—like those of
Xenocrates—rather than Stoic doctrine. (Not too much weight should be placed on
the phrase æ Ø ø å  itself; as Jonathan Barnes reminds me in correspondence,
it is just a natural expansion of the abbreviated phrase æ Ø.)
84
I am indebted for this discussion to conversations with Tobias Reinhardt and
Richard Sorabji; see Sorabji, Sourcebook III, 3(c) and Reinhardt 2007 for a helpful
summary of the issues. To fill out the inconsistency: (1) on the one hand, Simplicius at
least interprets Andronicus and Xenocrates as if they complained about the multitude
of ten categories (æØ e r ÆØ ŒÆ ’ ÆP f e F  H ªH ºBŁ , 63,22),
and encompassed the ten in two ( Æ fiH ŒÆŁ’ Æ e ŒÆd fiH æ Ø æغÆØ).
 Andronicus of Rhodes 51
tentative reconstruction. (i) Andronicus accepted that there were
substantial beings, which could properly be predicated per se in the
first category; (ii) he also accepted a broad sense of ‘relative’, which
could accommodate quantities, qualities, and perhaps most of the
other non-substance categories, and (iii) a narrow sense of ‘relative’,
which accommodated Aristotle’s strictly defined relatives and per-
haps also accommodated Andronicus’ rechristened categories of time
and place. These would break down in the following way.
(i) Beings in the first category. Andronicus accepted and inter-
preted a version of Aristotle’s story in Cat. ch. 2 about what
made an item a real, substantial being ( P Æ). Such a substan-
tial item could not be discovered residing IN any other item, in
the way that, say, paleness is IN Socrates’ body. Clearly Socrates
here (the primary P Æ) is not IN anything (in the technical
sense of Categories 1a24–5), and humanity (the secondary
P Æ) is also not IN anything. As a basic test, humanity (the
substance SAID-OF Socrates) cannot be removed from Socrates

Nonetheless, (2) Andronicus appears to preserve the number of ten categories (Simp-
lic. in Cat. 342,21–5). And (3) he declares that the relative is a kind of ‘offshoot’ or
‘sucker’ which belongs in order after all the other categories (in Cat. 157,18–22). How
will this evidence hang together? Moraux (1973: 103) suggests that Andronicus did
not intend to reduce the number of categories in this way, so that Simplicius is
representing him inaccurately or polemically, or else (107) that the evidence misrep-
resents Andronicus altogether, and he never posited such a bipartition. Huby (1981:
406) suggests that Andronicus allowed for two ‘supercategories’ of substance and
relation (and that he placed time under the supercategory of relation, as the measure
of movement, which would make him the natural target of Simplicius’ rebuttal to that
position at 134,25). Reinhardt (2007: 522) offers a nuanced alternative: Andronicus
may have (i) allowed for ten categories—one category of substance and nine distinct
non-substance categories—from a logical point of view, since kinds of properties are
logically or linguistically distinct from one another, and at the same time,
(ii) Andronicus may have collapsed all of the non-substance categories into one
from an ontological point of view, because for all of them ‘being amounts to existing
in relation to something else, namely the subject they are “in” ’. (Thus Andronicus
would have anticipated Boethus, further discussed in ch. 6, in distinguishing between
an ontological, epistemological, and linguistic interpretation of the Categories). Then
when Simplicius tells us of Andronicus’ reduction of the categories and shunting of
relation to the end of the list, he refers to his ontological view; when he tells us that he
maintained the list of ten, he refers to his linguistic view. (One challenge for this
interpretation, perhaps, is the lack of clear evidence for an ‘ontological’ reading of the
Categories offered by Andronicus as distinct from Boethus’ contribution; but it is
certainly possible that Andronicus’ view was adopted more or less wholesale by
Boethus in a revised form, and not preserved as distinct in the later tradition).
52 Andronicus on the Categories
without Socrates ceasing to exist, but paleness (the quality IN
Socrates) can be so removed, causing qualitative change in
Socrates, but not his disappearance.
(ii) Broad ‘relatives’: quantity, quality, and perhaps being-
in-a-position, having, doing, being-affected. But other,
non-substantial items are in a broad sense parasitic on P ÆØ:
this includes items in most of the remaining categories (pos-
sibly excluding time and place, Andronicus’ relabelled categor-
ies of when and where: see below). These are relative in a broad
sense (perhaps similar to the secundum dici account at
Cat. 6a36–7), spoken about primarily in their relation to
other things, viz. the subjects, the P ÆØ in which they inhere,
in a per accidens way. For example, it might seem impossible to
describe Socrates’ paleness without speaking of Socrates, or
colour in general without speaking of body in general.
(iii) Strict ‘relatives’: Relation, under which might also fall place
and time (and perhaps other categories).85 Finally, Andro-
nicus understood another kind of Relative (in a strict sense,
following his revised definition of the relative at Cat. 8a31) to
be an item whose entire being consisted in its disposition
toward something else (cf. Porph. in Cat. 125,14–23)—a
very ‘weak’ kind of being, comparable to the Stoic ‘category’
of the relatively disposed (æ Ø ø å ). Items that
were relative in this last, strict sense, such as (say) the prop-
erty of being on the right, could alter without anything intrin-
sically changing about the subject—unlike, above under (ii),
a proper quality such as paleness, whose alteration would
at least imply a significant change in the subject. On this
model, Andronicus did indeed reduce the number of ten
categories to two, insofar as he made all the nine non-
substance categories broadly ‘relative’ (perhaps following

85
Without complicating this picture too much, we may also wonder about the
place of the other, lesser categories, such as being-in-a-position and having: they may
fall under the ‘broad’ relative, as basically similar to  Ø, which is how I have
described them here, or possibly—since Andronicus’ scheme already answers to the
Stoic picture in parts—we might wonder whether he allowed some of these to answer
to the third Stoic ‘category’, the ø å  Æ or items disposed in certain ways, as
Simplicius himself describes ‘on the right’ and ‘to be shod’ (the latter an example of
‘having’ in Ar. Cat. 4, 1b24–2a4) as cases of ø å  Æ. See Menn 1999 on the Stoic
categories.
 Andronicus of Rhodes 53
Xenocrates). Yet he also maintained the number of ten (since
the nine non-substance categories are still distinct from one
another, although they are all in a loose way relative to
substance). Finally, he made the relative in the strict sense
properly posterior to substance and the other eight non-
substance categories, since it was a kind of ‘off-shoot’ or
‘appendage’.
It might be added that Andronicus may have regarded time as
relative in either a broad or strict sense, insofar as it is the measure of
movement (so Huby 1981: 406 suggests). If he treated time as a
subcategory of the relative in the strict sense, and moved time to the
end of the list of categories along with the strict relative, that might
have provided an impetus for the posterior position of time and place
at the end of the list of categories in other writers such as Philo of
Alexandria and ps.-Archytas (see ch. 3). There may also be some trace
of Stoic influence on Andronicus’ thinking here, since the Stoics
treated time as an incorporeal and thus less ‘real’ than corporeal
things; but that is pure speculation, and most arguments relating
to the order of the categories cannot be pressed too far: Aristotle
himself sometimes orders time at the end of the list (Metaph. ˜7,
1017a23–30).
If this interpretation is accurate, Andronicus’ exegesis of Aristotle
was basically different from that of the later Neoplatonist commen-
tators such as Simplicius, who take Aristotle to believe that one can
predicate in all the non-substance categories per se, except the relative
(cf. Simplic in Cat. 174,14–175,11; compare also Aristotle’s distinc-
tion of accidental and absolute predication at Metaph. ˜7,
1017a8–30). But as we will see (ch. 3), the first-century Academic
Eudorus will criticize Aristotle as if he treated all the non-substance
categories as relative. Eudorus may be responding to an interpretation
like Andronicus’, or Andronicus may be influenced by Eudorus, or
each may be independent of the other (see ch. 3).
We should also bear in mind the possibility that Andronicus was
encouraged by the text of Cat. ch. 7 to admit two rather different
accounts of the relative. One possibility, as Reinhardt tantalizingly
suggests (2007: 522), is that Andronicus was prepared to regard the
nine non-substance categories as ontologically identical (in that for
them, existence just is having a certain relationship to something else)
54 Andronicus on the Categories
yet logically or linguistically differentiable.86 This would be especially
intriguing if it implied that Andronicus anticipated the distinction
between a linguistic and ontological reading of the Categories that
would so preoccupy the later tradition.87 For our purposes here,
however, I have suggested only that Andronicus analyzed the Cat-
egories’ two accounts of the relative (6a36–7 and 8a31–2) as somehow
‘looser’ and ‘stricter’ senses,88 respectively applicable to the non-
substance categories in general and the category of relation itself,
in which he may have recognized (something like) the Stoic æ Ø
ø åØ, or a set of Academic categories used by Xenocrates or
Hermodorus.89

(c) The Categories, in introducing demonstration,

helps to avoid equivocation and fallacy
Returning to Andronicus’ catalogue and arguments for the place and
title of the Categories, we remember that Andronicus brought the
Categories to the fore of the curriculum and dissociated it from
merely ‘persuasive’ dialectic, emphasizing its value for demonstration
and the pursuit of truth. It would be attractive to speculate that
Andronicus saw in the core of the text (but not in the latter third of
it, the ‘post-praedicamenta’ beginning with ch. 10) a useful handbook
for resisting fallacies that turned on ambiguity and equivocation—
especially those that did not recognize the proper difference between
synonymous and homonymous predication. Thus Andronicus,
who was happy to jettison the last third of the treatise, carefully
maintained, paraphrased, and explained the first part, re-articulating

86
After all, Andronicus would hardly have denied that the nine non-substance
categories were logically distinct, but he could have denied that those distinctions
picked out any real difference between real beings.
87
But against that approach to the distinction in Aristotle himself, see Sedley 2002:
332–4. We might furnish some additional support to this interpretation from the two
definitions or descriptions of the relative offered by Aristotle in the Categories,
respectively ‘those which, whatever things they are themselves, are said to be of
other things, or in some other relation to another thing’ (6a36–7), and ‘those for
which being is the same as to be relatively disposed to something’ (8a31–2)—if we
were to understand these as respectively suggesting (to Andronicus’ ear) a de dicto and
de re account of relation.
88
Similar to David Sedley’s language (2002) of ‘soft’ and ‘hard’ relativity applied to
the passage in Aristotle, although I suggest that Andronicus’ interpretation is some-
what different.
89
On which see Sedley 2002: 348–51, Fine 1993: 176–82.
 Andronicus of Rhodes 55
the initial statement distinguishing homonyms from synonyms and
defending its importance for the entire treatise (cf. Simplicius in Cat.
21,21–4). Without the recipes that the Categories offers, we might also
fail to distinguish between homonymous and synonymous predica-
tion in general, as Boethius, perhaps channelling Andronicus by way
of Porphyry, stresses at De Div. 880C–D, 17,24–18,1 Magee.90 That
could make our definitions and demonstrations vulnerable to various
sophisms or lead us into mistaken metaphysics—errors like assuming
that every term that referred to many things picked out a universal
essence. Moreover, then, a work that began by setting out the distinc-
tion between homonymous and synonymous predication would
appear especially useful. This is the interpretation that I will attempt
to develop below.
The Categories might have been seen as useful for solving the kind
of puzzles that Cicero, speaking for Philo, levelled against Antiochus’
loyalty to dialectic. Of more direct relevance for us here, it might have
been seen as useful for the kinds of puzzles that Simplicius raised as
problematic for ‘the dialecticians’ who simply recommended ‘quies-
cence’—that is, for the Stoics—but soluble for the dialectician who
made use of the significata described in the Categories:
Some people . . . rightly say that it is from the realities (æªÆ Æ) that
homonyms become clear to us: viz., when the same name is spoken,
I project one concept in conjunction with the name, while you project a
different one. For example, if someone says the name ‘dog’, I might
conceive of the land-animal, while you might conceive of a sea-dog.
This is why the dialecticians recommend that we become quiescent
when faced with syllogisms based on homonymy, until the questioner
transfers the name to one of the significata . . . Moreover, Aristotle
himself presents his teaching on homonyms, because the discussion
(ºª ) of homonymy is also . . . immediately consequent upon the goal
( Π) of the Categories. For his part, Plato states that it becomes clear
from realities whether the same name is borne by different things
homonymously or synonymously, for he says in the Sophist, ‘for now,
you and I have the name in common with regard to this matter, but
perhaps we each privately keep the fact (æª ) that we are naming to
ourselves; [however, we ought always in every instance to come to
agreement about the thing itself by argument (Øa ºªø) rather than
about the mere name without argument.]’ (Soph. 218C). Thus, it is

90
For the pervasive uses of homonymy in Aristotle’s philosophy, see Shields 1999.
56 Andronicus on the Categories
when it becomes known that each individual has his own particular
 ØÆ of a common name that it becomes clear that the name is
homonymous . . . the nature of homonyms and synonyms is in the
 ØÆØ . . . of human beings as they converse and name things.91
(Simplicius in Cat. 24,6–25,9)
This explanation seems to suggest the value that the dialectician
should place on recognizing equivocation. It also implies that a list
of the significata, the summa genera, could be a helpful tool for
defeating some sophistical challenges, including some that troubled
the Stoics (cf. Aristotle, Soph. El. 22, 178a4–5, involving sophisms that
entail the confusion of one category for another, like  E  for  
or  Ø F for  å ).92 It might also contextualize the interest in
distinguishing accidental from essential predication attributed to
Andronicus in Boethius’ De Divisione, which also illustrates how
this distinction—including the distinction of differentiae per se and
per accidens—can be deployed to help determine what is univocal and
what is equivocal, a crucial aid in the construction of good definitions

91
 æ F Ø b ŒÆd Øa , ŒÆ Ø æd ºø æ Ł  NE, Iºº’ P æd ÆP H
H æƪ ø, ‹ø æd ›øıÆ b Pb ºªØ, æd b H ›øø
Ø ŒØ, ŒÆ Ø æ Ū ıÅ B ŒÆ a c ›øıÆ K Æ, Yæ I’ KŒÅ
a ›ıÆ. ŒÆd ºª ı Ø ŒÆºH, ‹ Ø Ie H æƪ ø ª ÆØ BºÆ a ›ıÆ,
‹ Æ F ÆP F OÆ  (10) ÞÅŁ  Kªg b ¼ººÅ  ØÆ, f b ¼ººÅ æd F
OÆ  æ ƺºŁÆ, u æ F Œø OÆ  ÞÅŁ  Kªg b e åæ ÆE , f
b e ŁÆº Ø  K ØÆ. Øe ŒÆd K E Ææ’ ›øıÆ ıºº ªØ  E  ıåÇØ
ƒ ØƺŒ ØŒ d ÆæÆŒº  ÆØ, "ø i K’ ¼ºº ÅÆØ  › Kæø H  ƪªfiÅ
e Z Æ· x , Y Ø Kæø fi A N › åØ g IæE , (15) N å Ø IæE  þ,
ıªåøæÅ ŁÆ· Œi Kæø fi Å N › IæE  hłıå , ŒÆd F ıªåøæÅ ŁÆ,
IºÅŁb ªæ· N b ıƪªfiÅ ‹ Ø › åØ g ¼æÆ hłıå , K ÆFŁÆ c ›øıÆ F
Iæ ı ØÆ ºÆ ŁÆØ ŒÆd EÆØ [ c IæÆ Xª ı c PłıåÆ] ‹ Ø ¼ººø b
Kd F åØ H , ¼ººø b Kd F c IæÆ å   ºª ÆØ. u  a æªÆ Æ
Œıæø, Påd a (20) OÆ Æ  ØE c ›øıÆ. ŒÆd ªaæ ÆP e Kd H ›øø
 ØE ÆØ c ØÆ ŒÆºÆ ıÆŒ º ıŁ F   ø fiH Œ fiH PŁf ŒÆd F æd B
›øıÆ ºª ı. ŒÆd › —º ø b KŒ H æƪ ø ź F ŁÆ çÅ Ø, Y 
›øø ŒÆ a ºØø çæ ÆØ e ÆP e Z Æ Y  ıøø· [p. 25] ºªØ ªaæ
K fiH  çØ fi B· “F ªaæ c f ŒIªg  ı æØ h Æ å  Œ ØB fi , e b æª , Kç’
fiz ŒÆº F, Œ æ  åÆ i NÆ fi Ææ’ E ÆP E å Ø”, u  ‹ Æ ªø ŁB fi ‹ Ø
F Œ ØB
fi ºª  ı OÆ  NÆ K d  ØÆ Ææ’ ŒÆ æø fi ,   ª ÆØ Bº , ‹ Ø
›ı  e Z Æ. P ªæ K Ø ÆP a Kç’ Æı H a ›ıÆ ŒÆd a ııÆ (5)
ØÆF Æ, Iºº’ K fi B H Øƺª ø ŒÆd O ÆÇ ø ÆæźºÆªfiÅ K Æ fi j
ıçø  fi Å  H ›øø ŒÆd ıøø K d ç Ø. ŒÆºH b e ºª ÆØ r ,
KØc c æd H æƪ ø ÆP H › ºª  K , Iººa æd H ÅÆ ØŒH ºø
(ed. Kalbfleisch 1907).
92
See also Menn 1995: 320.
 Andronicus of Rhodes 57
(De Div. 16,24–18,3 Magee). This is certainly an application of the
main division that Andronicus wanted to expound.
We might again deduce from this evidence that Andronicus valued
division as a tool serving demonstration, as we have seen above.
Certainly many of his contemporaries and successors did.93 Indeed,
Andronicus may have turned to Aristotelian resources after recog-
nizing a lack in Stoic logic, which Atherton points out:
a feature . . . conspicuous by its absence from the Stoic context, in com-
parison with the Platonic–Aristotelian: any systematic use of homonymy,
ambiguity, or related concepts as general instruments for identifying
and solving philosophical problems, or for analysing, constructing, and
disarming philosophical arguments. (Atherton 1993: 102)

(d) The Categories begins from pre-technical outlines

(  ªæÆçÆ ) of the genera that actuate our innate
preconceptions of them, making it a
suitable work for beginners
The association may be pressed further. Simplicius tells us that
Andronicus maintained that we require a æºÅłØ or ‘preconception’

93
Some examples are cited below. Aristotle himself diminishes the value of
division unlike Plato (An. Pr. 1.31, 46a31–2) and explicitly places limits on its
usefulness. Ryle agrees with Aristotle, for example in ‘Plato’s Parmenides II’, Mind
48 (1939), 302–25, reprinted in R.E. Allen (ed.), Studies in Plato’s Metaphysics
(London: Routledge & Kegan Paul, 1965), 97–147 and in Plato’s Progress (esp. 102
ff.) J. L. Ackrill, by contrast, offers a careful ‘Defence of Platonic Division’ (orig. pub.
1970) in Essays on Plato and Aristotle, Oxford: Clarendon Press 1997. For Aristotle,
famously (e.g. Cherniss 1944: ch. 1, Balme 1987), division does not ‘prove’ anything
(Pr. An. 46a31–9, An. Post. 91b12–15, 96b25–97a6). As Marguerite Deslauriers
stressed, Aristotle does value division, and he wants ‘to demonstrate the coherence
of the logical and ontological functions of form’ (1990: 216); cf. Deslauriers 2007:15 ff.
The following point may also be relevant: Simplicius credits to Andronicus (in Cat.
54,8) the observation that we predicate attributes of subjects transitively, in the sense
described by Aristotle in Cat. 1b10–15, not only in the first category or essentially (K
fiH  K Ø), but also in the other categories, as when we call a subject ‘cultured’ or
‘Athenian’: thus for example ‘Socrates is philosophical’, ‘philosophers are knowledge-
able’, so ‘Socrates is knowledgeable’ (Simplicius, in Cat. 54,8–21; cf. Moraux 1973,
104, and see the anonymous commentary in the Archimedes Palimpsest, 6,9–7,7
Chiaradonna, Rashed, and Sedley 2013, with commentary and references). As Sim-
plicius goes on to show, this kind of argument can lead to fallacies of equivocation that
turn on category errors: for instance, this body is white, and white is a colour;
therefore, this body is a colour. Andronicus might have thought that such inferences
turned on a failure to distinguish between accidental and essential predication (this
body is white per accidens, but white is a colour per se).
58 Andronicus on the Categories
about homonyms (more on this below) in order to grasp the core
function of the book Categories, and this is why the ‘onymies’—the
opening chapter’s account of homonyms (items that share only a
name in common, but not a definition), synonyms (items that share
both a name and a definition in common), and paronyms (items that
get their name from something with a difference of ending)—should
come first in the treatise (in Cat. 21,21–4). In a passage that derives
from Porphyry, Simplicius gives an example of how
. . . ‘man’ and ‘horse’ participate in the same substance, that of ‘animal’,
which is predicated of them synonymously, and they are therefore
placed under the same category. Socrates and the painted Socrates,
however, do not both participate in the substance of ‘animal’: but one
participates in substance, while the other participates in ‘colour’ or
‘surface figure’. Thus, they are not placed under the same category,
but Socrates comes under the category of Substance, while the painted
Socrates comes under the category of the Qualified.
It was thus necessary to give preliminary teaching about homonyms
and synonyms. The necessity of having a preliminary notion (æºÅłØ)
of homonyms was also shown by Andronicus, who made the initial
phrase of the Categories read as follows: ‘Of things said, some are said
without combination, others with combination. Of those without com-
bination, those which have the name alone in common are called
homonyms.’ Thus, a preliminary notion (æºÅłØ) of homonyms is
clearly necessary, both for the above-mentioned reasons, and because,
since there is a great deal of controversy about whether or not Being is a
genus, [the answer] can be known by means of homonyms and syn-
onyms.94 (21,15–22,1, tr. Chase 2003, lightly modified)
After Porphyry and Simplicius’ endorsement of Andronicus’ view
(and paraphrase), Simplicius adds that ‘there is nothing as effective
as a precise definition of names to counteract sophistical quibbling’

94
 ø ªaæ › b ¼Łæø  ŒÆd › ¥   B ÆP B P Æ F Çfi ı  å  
ıøø ÆP H ŒÆ Ū æ ıÅ e Æ Iª  ÆØ ŒÆ Ū æÆ, › b øŒæ Å
ŒÆd › ªªæÆ  –  c B P Æ ¼çø F Çfi ı  å  , Iºº’ › b B P Æ,
› b åæÆ  j å Æ  KØ ºÆ ı, På e c ÆP c Iª  ÆØ ŒÆ Ū æÆ, Iºº’
› b øŒæ Å e c P Æ, › b ªªæÆ  e e  Ø. Ø ı æ ØÆØ
æd F ›ø ı ŒÆd ıø ı. ‹ Ø b åæØÅ K d  H ›øø æºÅłØ,
K ºø  æØŒ  æ Æ “ H ºª ø a b ¼ı ıº ŒB ºª ÆØ, a b
 a ıº ŒB· ŒÆd H ¼ı ıº ŒB ›ıÆ b ºª ÆØ, z Z Æ  
Œ Ø”. u  åæØÅ  H ›øø çÆ ÆØ æºÅłØ Ø  a NæÅÆ ŒÆd
‹ Ø  ººB h Å IçØ ºÆ, Y  ª  e k Y   , KŒ H ›øø ŒÆd
ıøø ªØ Œ ÆØ (ed. Kalbfleisch 1907).
 Andronicus of Rhodes 59
(22,10-11), followed by a Speusippan division of homonymy, poly-
onomy, synonymy, heteronymy, and paronymy, and the ‘manly
cloak’ example cited above—already familiar from Stoic dialectic
and the challenges levelled against it. Similar challenges were later
cited by Nicostratus against the Categories (specifically, against the
‘transitivity principle’ of 1b10–11; see Chiaradonna, Rashed, Sedley,
and Tchernetska 2013: 1,26–3,1).
Simplicius here preserves Porphyry, who perhaps preserves in his
turn Andronicus’ own argument for the usefulness of a sketch at the
outset of the Categories—introducing the notions of equivocal and
univocal predication by excavating a ‘preconception’ or æºÅłØ of
them. (I take it that Porphyry did not merely read Andronicus’
paraphrase of the opening of Categories and redeploy it as a proof
text for his own reasoning about the value of the ‘onymies’, but that
he built on an argument by Andronicus. Thus in 379,8–11 below
I think that we can see traces of Porphyry’s response to an earlier
Andronican argument for the value of the ‘onymies’, and Herminus
[ap. Porph. in Cat. 59,17–33], who is there ‘compressing’ Andronicus’
pupil Boethus of Sidon, demonstrates that the argument for the
Categories as useful for teasing out and correcting æ º łØ is pre-
Porphyrian and dates to the first century.)
What notion of æºÅłØ could Andronicus build upon (if indeed
he uses the word in a technical sense here)? In a broad Epicurean and
Stoic usage, a æºÅłØ is a rough preconception that arises naturally
in the course of the use of our senses: in Epicurean usage, for instance,
repeated sense-impressions of a dog will generate the preconception
‘dog’, which is a kind of rough, natural sketch of what we have
grasped about dogs (NºÅçÆØ) prior to (æ) a scientific or precise
inquiry.95 Based on this—as the Neoplatonists, too, agreed96—we are
able to bootstrap a process of conversation or dialectic. For the Stoics,
on one interpretation (Brittain 2005: 179), some of these preconcep-
tions are common conceptions or Œ ØÆd  ØÆØ—or they become
common conceptions once they are properly excavated or articulated
(Dyson 2009: 53–71)—and it takes a process of working from a rough
sketch or  ªæÆç toward a precise definition to ‘excavate’ their

95
See Asmis 1984: 22 for this suggested etymology, and Brittain 2005 and Dyson
2009 for the Stoic and Epicurean usage more broadly.
96
For the Neoplatonic reception of æ º łØ following Plotinus, see Van Den
Berg 2009 and Helmig 2012: 141–342.
60 Andronicus on the Categories
content. This process of excavation is called ØæŁæø Ø or ‘articula-
tion’ (for which see for example Helmig 2012: 280, Brittain 2005:
186): a commonplace of Stoic, Platonist, and other traditions (com-
pare for example Anon. in Theaet. col. 46, 43–9), it involves shifting
from ordinary linguistic usage, our day-to-day language, toward a
more precise, clearly defined usage that has been hardened by philo-
sophical articulation (cf. Brittain 2005: 157).
Did Andronicus have some such Stoic infrastructure in mind? That
he did is implicit, I think, in a response later offered by Porphyry
against Andronicus’ disconnection of chs. 10–15, the so-called ‘post-
praedicamenta’, from the rest of the Categories (mentioned by
Simplicius at 379,8–11). Andronicus should not have rejected that
part of the treatise, Porphyry explains, because he ought to have
recognized that these later sketches of opposition, priority and pos-
teriority, and so on were necessary for the same reason as the onymies
at the beginning of the treatise:
Others, including Porphyry, opine that these investigations [i.e. the
post-praedicamenta] contribute to [the] clarity [of the work], because,
among the names which receive mention in the Categories, those which
were not anticipated (æ غÅÆ) by the common conceptions
(Œ Øa K Æ) Aristotle took right at the beginning and articulated
(Ø æŁæø ) them, as for example in the account concerning hom-
onymous and synonymous things, while those which were indeed
anticipated in the common conceptions, but which required further
articulation (ØÆæŁæ ø)—of these Aristotle analysed [in each case]
the confused [name] belonging to the preconception (æ º łø) after
the completion [of the Categories proper], in order not to chop up the
continuity of the account by inserting the articulation (ØæŁæø Ø) of
these [names] into the midst [of the main account].97 (Simplicius in
Cat. 379,12–20, tr. Chase 2003).98

97
¼ºº Ø b ŒÆd — æçıæø
fi Iæ ŒØ æe Æç ØÆ ı ºE a Łøæ Æ Æ ÆF Æ,
Ø Ø H K ÆE ŒÆ Ū æÆØ  Å ıå ø O  ø, ‹ Æ b PŒ q ŒÆ a a
Œ Øa K Æ æ غÅÆ, ÆF Æ K IæåB
fi æ hºÆ › æØ ºÅ ŒÆd Ø æŁæø ,
u æ e æd H ›øø ŒÆd ıøø ºª , ‹ Æ b q b K ÆE Œ ØÆE
K ÆØ æ غÅÆ, ØÆæŁæ ø b º   KE ,  ø e ıªŒåı  B
æ º łø  a c ıº æø Ø Ø æŁæø  æe e c ØÆŒ Ø e F ºª ı
ıåb  Æf c  ø ØæŁæø Ø Æ溺  Æ. ƒ b ÆF Æ ºª   ºª ı Ø
 Ø, P  Ø ª B ‹ºÅ ÆN Æ æØŒæÆ F Ø (ed. Kalbfleisch 1907).
98
Compare also Philoponus in Cat. 167,21–168,3 and Ammonius in Cat. 93,9–12.
 Andronicus of Rhodes 61
I would interpret Porphyry as here endorsing an Andronican argument
that the Categories begins—in the first third, or prepraedicamenta—by
capturing and teasing out a æºÅłØ of equivocation, of which we lack
a common notion.99 Andronicus’ error is in failing to recognize that
the Categories does just the same thing in the last third, or post-
praedicamenta, this time focusing on teasing out and excavating our
æ º łØ of things of which we have a common notion, on the basis of
which we were able to read the text. That would suggest that Andro-
nicus had some notion of ‘articulation’ (ØæŁæø Ø) along the lines of
the Stoic excavation of preconceptions, and Porphyry, while accepting
his basic premises, thinks that he has failed to apply it to the post-
praedicamenta as he applied it to the pre-praedicamenta or onymies
(above, 21,15—22,1).
There is another link between Porphyry and Andronicus on the
methodology of the ‘articulation’ of concepts through dialectic that
suggests this. In his discussion of quantity, Simplicius comments that:
Porphyry tries, on this question, to follow Andronicus in articulating
the concept (ØÆæŁæ F . . . c  ØÆ) of the one, the unit and the
point, not when spoken of in terms of Ideal Forms but when evident
in terms of perception or reasoning (which the present undertaking is
concerned with), let us too try to follow what he says.100 (154,3—6, tr.
Fleet in De Haas and Fleet 2001)
The ‘articulation’ follows. Porphyry begins from a perceptible, every-
day example, the unity of a body (154,7—12):
When we say that a length or a breadth or a depth is one, we must
consider in what way we mean ‘one’. Now it is clear that when someone
looks at a body which is a continuous whole in itself containing all its
parts, but is circumscribed individually and separated from all other
bodies, we call such a continuous body ‘one’ and the continuity of its
parts with each other ‘a unity’. In this way we conceive of what is
numerically one.101

99
For Porphyry’s reception of earlier, Hellenistic concepts of prolēpsis and ennoia,
see Chase 2010 and 2000: ch. 1.
100
! ¯ Øc b K ÆFŁÆ ØÆæŁæ F › — æçæØ  KØåØæE fiH æ Œø fi
ŒÆ ÆŒ º ıŁH c  ØÆ F  e ŒÆd B    ŒÆd B تB, P H ŒÆ ’
NÆ ºª ø, Iººa H ŒÆ ’ ÆY ŁÅ Ø j Ø ØÆ KÆæªH, æd z ŒÆd  Ææ F Æ
æŁ Ø æƪÆ  ÆØ, çæ ŒÆd E E ºª  Ø ÆæÆŒ º ıŁ ø (ed.
Kalbfleisch 1907).
101
‹ Æ ªæ, çÅ , £ ºªø j BŒ  j º  j Ł , Œ   H e "
çÆ. j Bº  ‰ › Æ Ø HÆ Ł Å ÆØ ıåb b ‹º  Æı fiH æd  Æ a
62 Andronicus on the Categories
‘Pumping our intuition’ of unity in this fairly ordinary way, which
requires no technical expertise, he proceeds to offer a more technical
example (154,12—13):
[W]e recognize what is one in species and of the same species <as other
things> by separating off what is accidental to the individuals according
to the differentiae in them.102
Porphyry’s ‘articulation’ of the  ØÆ of unity here is no doubt a
commonplace, and nothing special; but Simplicius says that he has it
from Andronicus, which presumably implies that Porphyry himself
credited this particular ‘articulation’ to Andronicus. It might illustrate
just the kind of ‘excavation’ of an  ØÆ that Andronicus deployed
the Categories to facilitate. That is, Andronicus might have suggested
that the Categories, with its relatively non-technical delineations of
the highest genera, does for our innate concepts of ‘relation’ or
‘quantity’ something like what he does here for our concept of
‘unity’.103 (If this is right, we should also note that Andronicus
sometimes tries to improve on the treatise, as when he amends the
wording of the Categories’ account of the æ Ø [Simplic. in Cat.
202,5].)
I think that we can strengthen this interpretation from other
sources, tracing this approach to the Categories to Herminus in the
second century ce, who in turn follows Boethus of Sidon in the first
century. To develop this case, we now return to the argument in
favour of the title Categories preserved in Porphyry’s shorter com-
mentary. For a long stretch in the opening pages of the commentary,
Porphyry contends that the treatise should not be called Before the
Topics but instead ˚Æ ª æÆØ, because it is about ŒÆ Ū æÆØ, simple
significant expressions. This argument about the title, as I proposed
above, should likely be attributed to Andronicus, on the strength of
coordinate passages like Simplicius in Cat. 379,8–11. In Porphyry’s
commentary, it is backed by a variety of stronger points about the
nature of language acquisition grounded, not in Aristotle, but in Stoic

æØÆ, æتªæÆ  b NÆfi ŒÆd H ¼ººø ± ø ŒåøæØ  , £ e Ø F 

ıåb O Ç  ŒÆd c H  æø ÆP F æe ¼ººÅºÆ ıåØÆ "ø Ø. ŒÆd
o ø e fiH IæØŁfiH £  F . . .
102
e b fiH YØ £ ŒÆd › Øb ªØ Œ  a ıÅŒ Æ E I  Ø ŒÆ a
a K  Æ ØÆç æa åøæÇ  .
103
Compare 26,17 for Andronicus’ use of a concept of unity.
 Andronicus of Rhodes 63
and Epicurean philosophy of language,104 which I think may be a
hallmark of Andronicus’ approach. (Boethus, on the other hand,
shows signs of adopting the semantics of De Interpretatione as a
new framework for interpreting the Categories; see ch. 6.)
Porphyry states that everyday language lacks precise names for
certain real things dealt with by philosophy, but the Categories can aid
us in acquiring a new vocabulary (cf. Porph. in Cat. 55,3, Simplicius
in Cat. 74,4), thereby bringing our language closer to reality. The
Categories requires just a few coined or ‘technical’ terms to introduce
philosophy properly (ºØ ŒÆØÆ, for which cf. Porphyry in Cat.
55,12 and the ‘five words’ of Isag. 1,4–5). Porphyry credits the general
thrust of his argumentation to Boethus of Sidon, and then adds
that Herminus (the second-century ce Peripatetic who lectured to
Alexander) takes the same view as Boethus and himself, namely that:
[T]he subject of the [Categories] is not the primary and highest genera
in nature, for instruction in these is not suitable for young persons, nor
the issue of what the primary and fundamental differentiae of things
said are, since in that case the discussion would seem to be about the
parts of speech. Rather it is about the sort of predication that will
properly belong to what is said in each of the genera of being. Hence
it also became necessary to touch in some way upon the genera to which
the predications in question correspond, for it is impossible to recognize
the kind of signification that is proper to each genus without some
preconception (æºÅłØ) of it. This also accounts for the title Predica-
tion (ŒÆ Ū æÆ), which means ‘the proper mode of signification con-
nected with each genus’.105 (Herminus ap. Porphyrium in Cat.
59,17–33, tr. Strange 1992, lightly modified)

104
See Frede and Inwood 2005: 14–54; I have also tried to explore this question
further in Griffin 2012b.
105
¸ªØ ı ›  EæE  æ ŒE ŁÆØ h  æd H K fi B ç Ø æ ø ŒÆd
ªØŒø  ø ªH ( P ªaæ  Ø æ Œ ı Æ  H Ø  ø ØÆ ŒÆºÆ) h  
ƃ æH ÆØ ŒÆd ØåØØ H ºª ø ØÆç æÆ, ‰ e ºª  r ÆØ  ŒE æd
H F ºª ı æH, Iººa Aºº  æd B ŒÆŁ’ "ŒÆ  ª  H Z ø NŒÆ i
K Å H ºª ø ŒÆ Ū æÆ· Øe ŒÆd IƪŒÆE  Kª Iø ªø –łÆ ŁÆØ
H ªH, Kç’ –æ  H ŒÆ Ū æ ıø IÆç æ· IÆ  ªaæ c Œ ı
NŒÆ ÅÆ Æ ªæØ  r ÆØ ÅØA ÆP fiH æ º łø æ ı ŒØÅ. Øa
F b ŒÆd  KتæÆçc ˚Æ Åª æÆ, źø ØŒc s Æ B Œ fi ø ªØ ıÅÅ
NŒÆ ÅÆ Æ. ‹ Ø b ÆP a a ªÅ e IæØŁe ŒÆ, æ œg › ºª  ź Ø·
Æ fi Å b ŒÆd › IæØŁe › H ŒÆ Ū æØH ŒÆ. › b Kتæçø —æd ŒÆ ªH Pb
ÆP e IØŒ ø, N   æe c Kd a ªÅ IÆç æa  Ø E c KتæÆç , Iººa
c ‹ Ø æ Ū ıø æd H ŒÆ ªH  Ç Ø (ed. Bodéüs 2008).
64 Andronicus on the Categories
This is evidently a compressed version of Boethus’ interpretation
(BÅŁ  K E N a ˚Æ Åª æÆ YæÅŒ ÆF Æ ŒÆd  EæE 
æÆåø, 59,17–18). It seems to represent a similar reading of the
Categories as a text to be used for teasing out or articulating
æ º łØ. Perhaps most importantly, it is extremely suggestive of a
justification for beginning the philosophical curriculum with the
Categories, because this treatise is uniquely well adapted to help to
articulate precisely these æ º łØ. A similar view is attributed to
Boethus in Simplicius’ commentary:
It was not feasible to give definitions (‹æ Ø) of the primary genera . . . But
it was possible, by means of a general sketch account (  ªæÆç ), to
actuate (IÆŒØE) our conception ( ØÆ) that fits ( ıÆæÇ ı Æ)
with the relative. [Aristotle] does this by following Plato according to
the first rendering (I Ø) [of the definition of the relative], as
Boethus [of Sidon] tells us. For Plato is said by him to have given the
following rendering about relatives: ‘whatever are said to be just what
they are [as being] of other things’.106 (Simplicius 159,10–15, transla-
tion Fleet 2002, lightly modified)
I take this passage to suggest that Boethus, for his part, located
the Categories’ value in the core chapters’ ‘sketches’ or  ªæÆçÆ of
the genera, stirring up our concepts about them without requiring the
capacity to construct a technical definition (which would also be
impossible, in the absence of any higher genera). (The verb IÆŒØE
has a storied philosophical history in the Platonic tradition,107 but
that does not prevent Boethus from having used it in commenting on
Aristotle; at any rate, the following passage shows that the central
point about  ªæÆçÆ is Boethus’ own.) Another useful passage in
this connection, I think, is Simplicius in Cat. 163,28–9. Boethus
remarks that the first account of the relative (Cat. 6a36–7) looks
circular, which is no good in a definition; Simplicius rebukes him

106
! ¯ d b c ºØ N  ºª  ‹ Ø ‹æ  b I ØÆØ H æ Ø IÆ 
q· H ªaæ æ ø ªH ‹æ ı I ØÆØ I åÆ  q Øa a NæÅÆ æ æ 
ÆN Æ· Ø’  ªæÆçB  Ø  c  æÆ  ØÆ IÆŒØE c ıÆæÇ ı Æ E
æ Ø ıÆ e q. ŒÆd F  ØE fiH —º øØ ŒÆ a c I Ø c æ Å
IŒ º ıŁH, u çÅ Ø BÅŁ · ºª ÆØ ªaæ ŒÆd › —º ø o ø ’ ÆP F I  FÆØ
æd H æ Ø· ‘‹ Æ ÆP a –æ K d  æø ºª ÆØ’ (ed. Kalbfleisch 1907).
107
See Plato, Meno 85C, and for the Platonic context, Chiaradonna 2007c. It is
certainly possible that Boethus did not use this exact verb, but it is paraphrased by
Simplicius or his source.
 Andronicus of Rhodes 65
for failing to recognize that Aristotle is not offering a proper defin-
ition, but a sketch account or  ªæÆç . The rebuke stings because:
[Boethus] himself goes on to claim in his defence that it is necessary to
present sketch accounts (  ªæÆça) of the primary genera by means of
the things which are posterior to them as well as [by means of]
themselves.108 (in Cat. 163,28–9, tr. Fleet 2002, lightly modified)
Boethus stressed that the primary genera should be described by
 ªæÆçÆ. (It would, after all, appear to be impossible to furnish a
formal Aristotelian definition of the highest genera, since a formal
definition would comprise a genus and differences.)109 All of these
passages suggest that this broad approach to the Categories belongs to
the first century and to the first generations of interest in the treatise.
Even if this view originated with Boethus, I think it is a good
candidate for elucidating the interest in the Categories among the
earliest generations of commentators. And as I began to argue above,
I think there is also some circumstantial evidence that Boethus was
here following Andronicus.110
On the Stoic view, the difficulty about using such ‘sketches’ to
actuate preconceptions from ordinary language was that they capture
only accidental, not essential, features of the world (Brittain 2005:
196–200—like the merely ‘ennoematic’ definitions described by
Galen, Diff. puls. 4, 708,16–709,5). To arrive at an essential account
that was not merely expressed in terms of accidental attributes—that
was not a mere sketch but a proper definition—would require, for the
Stoic, a systematic dialectical process of articulation (ØæŁæø Ø), as
noted above. The Categories, I think, is construed as bridging this gap.
Consider, for example, Simplicius’ preface to Aristotle’s description of
the first category, P Æ:
Earlier <Aristotle> enumerated all genera in order, and he provided an
outline elucidation (  ıø ØŒc . . . ØÆ ŒÆºÆ) of them by means
of examples; now he takes in hand each of the genera and gives a more
detailed account as best as one can, by pointing out their conceptual

108
ŒÆd ÆP e b æ ºŁg I º ªE ÆØ ‹ Ø H æ ø ªH IªŒÅ Øa H
æø ŒÆd Ø’ ÆP H a  ªæÆça I ØÆØ (ed. Kalbfleisch 1907).
109
I am indebted to Marwan Rashed for this point: see Alexander, Quaestio De
differentia II [12!](a), translated in Rashed 2007: 64–5.
110
It might also be worth noting that the rendering preferred as a  ªæÆç here
by Boethus is close to Andronicus’ and Ariston’s preferred account of the relative as
disposed to another (Simplic. in Cat. 202,5).
66 Andronicus on the Categories
content by means of a sketch account ( ØÆ Ø’  ªæÆçB
KØØŒø), and by clarifying their attributes (ÆæÆŒ º ıŁ F Æ) and
distinguishing characteristics (YØÆ). Archytas proceeded in this way as
well.111 (Simplic. in Cat. 75,28–31, tr. de Haas and Fleet 2001, lightly
modified)
Here, again, is the fundamental pedagogical value attributed to the
Categories: Aristotle begins from relatively informal, everyday, percep-
tible ‘examples’ of the genera and then clarifies their ÆæÆŒ º ıŁ F Æ
and YØÆ, not requiring prior experience in the technicalities of con-
structing definitions. Thus Aristotle’s Categories simply offers an articu-
lation of familiar terms, approaching a more exact understanding:
Aristotle’s aim here is to teach us about the words mentioned in the
doctrine of categories that require some articulation (ØæŁæø Ø) and
explanation.112 (Ammonius in Cat. 93,9–12, tr. Cohen and Matthews
1991)
The Neoplatonists will later adapt this basic Aristotelian notion—that
the Categories is beginning from the familiar and proceeding to the
unfamiliar, from what is known to us to what is known by nature—to
accommodate a doctrinally valuable restriction of the Categories’
scope, establishing that it concerns itself only with sensible being.
Thus Simplicius contends that:
[Aristotle in the Categories] is talking about perceptible things
(ÆN ŁÅ ), which are also what is investigated by the ordinary person
(›  º ¼Łæø ) . . . he seeks the difference in accordance with those
meaningful words which were first and most properly assigned to
sensible things, and which are familiar to the ordinary person.113
(Simplicius, in Cat. 74,3–17)

111
—æ æ  b K Ø  Æ a ªÅ ŒÆ ÅæØŁ Æ ,  ıø ØŒc ÆP H c
ØÆ ŒÆºÆ Øa ÆæÆت ø  ØÅ  · F b "ŒÆ  æ åØæØÇ  c
IŒæØ æÆ ‰ x   Ææ Ø ÆP F  ØE ÆØ,    ØÆ Ø’  ªæÆçB
KØØŒø ŒÆd a ÆæÆŒ º ıŁ F Æ ŒÆd a YØÆ IÆŒÆŁÆØæ . o ø b ŒÆd
æå Æ K Å  . . . (ed. Kalbfleisch 1907).
112
Œ  K Ø K ÆFŁÆ fiH æØ ºØ ØÆØ A æd H çøH, z K fi B
ØÆ ŒÆºÆfi H ŒÆ Ū æØH KÅı , P Æ Æ Ø b E Iª ø P H KŒ
B ıÅŁÆ,  ø b ‹ø Øe ØÆæŁæ ø ŒÆd ØÆ ŒÆºÆ (ed. Busse 1895).
113
Kd ‹ Ø æd H ÆN ŁÅ H Øƺª ÆØ, æd z ŒÆd ›  ºf ¼Łæø  c
K ŒłØ  ØE ÆØ, Bº K Ø æH  b KŒ F c Ææa A Ø ºª Å P Æ
æ åØæ Æ ŁÆØ, Ø Æ K z ıE P ÆØ Å  Æ, ÆN ŁÅ B  ŒÆd ØÆ Å B,
N b c bæ Æ Æ Åb IºŁg ‹ºø c ÆN ŁÅ c ŒıæØø æÆ r ÆØ B
ØÆ Å B çÅ Ø, ‰ K fiH æd H ÆN ŁÅ H ŒÆ Ø æ Ū   å ı Æ
 Andronicus of Rhodes 67
[Aristotle in this treatise] is not concerned with discussing the intelli-
gible substances . . . this he made clear by the <word> ‘called’ (ºª Å,
Cat. 2a12). For in ordinary language ( ı ŁØÆ) intelligible substance is
not spoken of, nor is it known to the multitude ( ƒ  ºº ), but sensible
substance is.114 (Simplicius, in Cat. 82,2–20)
In fact, Boethus of Sidon had already developed the basic argument
that the Categories is concerned with the sensible world addressed by
speech, in response to criticisms raised by a predecessor of Nicostra-
tus and Plotinus (ap. Simplic. in Cat. 78,4–20: perhaps Eudorus, as
Chiaradonna 2009: 104 suggests, or Lucius).115 (For Plotinus, on the
other hand, language primarily addresses the intelligible;116 I argue in
ch. 3 that Eudorus of Alexandria may also have maintained that
language addresses both intelligible and sensible realms.)
I think this is a suggestive bundle of evidence supporting an early
reading of the Categories as ‘articulating’ æ º łØ in something
much like the Stoic sense, also illustrating and grounding its forward
position in the curriculum as the ‘first book’ of philosophy and logic.
The Categories alone offers the ‘honing’ of our natural preconceptions
of the highest genera that the later Aristotelian curriculum—and
dialectic in general—depend upon: as a dialectical handbook, it offers
and describes a short list of summa genera, including simple linguistic
‘tests’ that we can use to help determine to which category a given
term really belongs. (As I suggested above, this might also have been
emphasized as an important practical step in tackling sophistical

ºª , K fiz ŒÆd  ØÆ Å c P Æ P ŒÆ a c ÆP B ç Ø ŁøæE ÆØ, Iººa ŒÆ a c

å Ø, m åØ æe c ÆN ŁÅ , (10) ŒÆŁ’ m ŒÆd E Ie b B ÆN ŁÅ B N c
ØÆ Å c IÆÆØ ıŁÆ, Ie b B ØÆ Å B c ÆN ŁÅ c ŒÆ Ø  ººB
IØæÆ ŒÆd I æØ Æ IƺŠŁE Æ ‹æø fi æغÆ . P   b Kd B
P Æ, Iººa ŒÆd Kd H ¼ººø ŒÆ Ū æØH a ÆN ŁÅ a ŒÆd æØŒa H ŒÆŁº ı ŒÆd
ØÆ Å H æ  Ø, ŒÆd F NŒ ø  ØE, ŒÆ a a ÅÆ ØŒa (15) çøa c
ØÆç æa  Ø, ÆQ æ ø ŒÆd Œıæø Kd H ÆN ŁÅ H K åŁÅ Æ ŒÆd E  ºº E
IŁæ Ø N d ªæØ Ø . . . (ed. Kalbfleisch 1907)
114
j ‹ Ø P æŒØ ÆØ ÆP fiH æd H  Å H P ØH F NE, Iººa H ÆN ŁÅ H
c ç Ø ŒÆd Ø ÆæÆ FÆØ, ‹æ ŒÆd Øa F ºª Å KÆ · P ªaæ ºª ÆØ
K fi B ıÅŁÆfi Pb ªø ÆØ E  ºº E KŒÅ, Iºº’ Æo Å Aºº   ÆN ŁÅ (ed.
Kalbfleisch 1907).
115
See ch. 6.
116
See Chiaradonna 2002: 263–7; van den Berg 2008: 72 (referencing Dexippus in
Cat. 41,25–30).
68 Andronicus on the Categories
challenges that turn on equivocation).117 I think the attribution of
some such interpretation of the Categories to the first generation of
commentators, especially Boethus, is fairly strong, on the basis of the
evidence treated here.
I think that there is further reason to trace it to Andronicus, given
his apparent innovation in foregrounding the Categories in the corpus
of Aristotle, his interest in distinguishing accidental from essential
predication (as preserved in Boethius), and his associations with the
terminology of æ º łØ and ØæŁæø Ø in Simplicius. The Categor-
ies will activate our preconceptions of the highest genera and also
correct them, and—crucially—will begin us on the road to using our
honed concepts in scientific demonstration or IØØ, not merely
in the rhetorical applications of ‘dialectic’ in the Topics’ sense. The use
of the Categories’ simple tests helps us to avoid the mistake of
confusing accidental with essential predication, and so trains the
budding dialectician to resist sophistical puzzles turning on equivo-
cation (and such puzzles, to judge from Cicero’s criticisms, may still
have troubled Antiochus’ school before Andronicus). But it will not
be enough for this process simply to apply in rhetorical contexts:
philosophers would merely be ‘declaiming Ł Ø’ or rhetorical theses
maintained in dialectical practice (which, interestingly, is just what
Strabo says they were limited to doing before the recovery of the
esoteric treatises: Geog. 13.1.54). This is what Peripatetics and Aca-
demics had done before (cf. Brittain 2001: 329–42 and Reinhardt
2003: 347).
Andronicus perhaps argued that Aristotle provided the resources
that the ‘dogmatist’ required to articulate our innate  ØÆØ—a long
road that began with the Categories’ (pre-technical) training in the
predication of simple terms, and ran through propositions and syl-
logisms to the Analytics and IØØ. Our everyday speech, based
on our words’ conventional relation to everyday things, still encodes
something of the truth about things, a truth that Aristotle’s articula-
tions help to decode.118 The proof that predication is possible—that
words could successfully refer to things, once we are able to bring

117
Topics 1.9 and Soph. El. 22 presuppose such tests, as Menn 1995: 320 points out,
but do not provide the tools to practise them; the Categories does provide just such
tests.
118
The later Neoplatonist commentators would develop such a reading of its value:
see Simplicius in Cat. 12,10–13,12 with Hoffmann 1987. I have also explored some
issues relating to the Neoplatonists’ pedagogical use of the Categories in Griffin 2013a.
 Andronicus of Rhodes 69
every term under the correct genus—would lie in our subsequent
agreement in employing the predications that the Categories
articulated.119
Be that as it may, Andronicus certainly insisted that the Categories
contributes to demonstrative knowledge (IØØ) and not merely
to rhetorical competence, and in this he was not alone. There are
many traces of early interest in the value of formal division and
definition for finding the truth, focusing on the indefiniteness or
‘uncountability’ of particulars.120 Antiochus already crafts an ‘Old
Academic’ consensus that definition and division can be used to
overcome the vagueness of sensible reality (Cic. Ac. 1.5, above). The
Categories is represented as a solution to that same ‘vagueness’ of the
perceptible world in Porphyry’s shorter commentary on the Categor-
ies, building on the earlier discussion of the appropriateness of the
title ˚Æ Åª æÆØ:
For things and expressions are both practically infinite in number. But
his intention is not to list expressions one by one—for each one signifies
one particular being—but since things that are many in number are one
in species or in genus, the infinity of beings and of the expressions that
signify them is found to be included under a list of ten genera. Since
beings are comprehended by ten generic sorts, the words that indicate
them have also come to be ten in genus, and are themselves also so
classified. Thus predications (ŒÆ Ū æÆØ) are said to be ten in genus,
just as beings themselves are ten in genus.121 (Porph. in Cat. 58,5—13,
tr. Strange 1992, lightly modified)
The most general items, then, are ten; the most special are of a certain
number, but a definite one; the individuals—that is to say, the items
after the most special items—are infinite. That is why Plato advised
those who descend from the most general items to the most special to

119
For that kind of argument, an argument from our consensus about (properly
articulated) Œ ØÆd  ØÆØ, see for example Obbink 1992, Brittain 2005 (with passages
cited above).
120
I have also argued this point in Griffin 2012a and 2013b. For the Neoplato-
nist reception of this idea, see Chase 2011.
121
¼ØæÆ b <ªÆæ> åe ŒÆd a æªÆ Æ ŒÆd ƃ ºØ ŒÆ a IæØŁ. Iºº’ P
a ŒÆ a IæØŁe æŒØ ÆØ ØºŁE ºØ· Œ Å ªaæ ŒÆ a IæØŁe<"> ÅÆØ
H Z ø· Iºº’ Kd fi H IæØŁfiH  ºº K Ø £ Z Æ fiH YØ j fi H ªØ, ŒÆd  IØæÆ
H Z ø ŒÆd H ÅÆØ ı H ÆP a ºø N ŒÆ ªÅ oæÅ ÆØ æغÆÆ Å
N e ªæç ŁÆØ. N ŒÆ ı ªØŒa ØÆç æa æغÅçŁ ø H Z ø ŒÆ ŒÆd
ƃ ź F ÆØ ÆF Æ çøÆd ªªÆ Ø ŒÆ a ªÅ ŒÆd ÆP Æd æغÅçŁE ÆØ. ŒÆ s
ºª  ÆØ ŒÆ Ū æÆØ fiH ªØ ź  Ø, u æ ŒÆd ÆP a a Z Æ ŒÆ fiH ªØ (ed.
Bodéüs 2008).
70 Andronicus on the Categories
stop there, and to descend through the intermediates . . . and he tells us
to leave the infinites alone, for there will be no knowledge of them.
So when we are descending to the most special items, it is necessary to
divide and to proceed through a plurality, and when we are ascending
to the most general items, it is necessary to bring the plurality together
into one.122 (Porph. Isagoge 6,10—16, tr. Barnes, modified)
Alexander of Aphrodisias similarly, in commenting on Topics 1.12,
(in Top. 86,25–7), suggests that the difficulty about getting to know
the infinite particulars lies in our inability to ‘work through’
(IØØ Å Æ) them ‘part by part’ (ŒÆ a æ ), itself a nice echo of
An. Post. 1.22. Alcinous has a similar view (Didasc. 25, 3.6). And as we
have discussed, this approach to articulating the value of the Categor-
ies seems likely to be traced, through Herminus, to Boethus of Sidon,
and perhaps earlier to Andronicus. Pseudo-Archytas—writing per-
haps in the first century bce123—argues that:
Beings are spoken of in two ways: for some are subjects, others inherent
in subjects or accidental to them [ . . . ] And it is clear that P Æ and
 Ø Å and   Å, and the other thoughts and signifiers [mentioned
previously in the list of ten ŒÆ Ū æÆØ at 22,14] will have been spoken
of in just the same number of ways [ . . . ] All knowledge, then, takes its
starting-points (Iæå Å) from the limited (æÆ ø) and gets to
know the unlimited (Iæø); and this is even more true of the
knowledge of beings [ . . . ] Every craft and every form of knowledge
has a certain rank and definition, and anything like this is in a numerical
sequence (IæØŁe): and the entirety of number is ten . . . Now other
men have language (ºª ) by accident (for they use it without know-
ledge), but the wise have it in an absolute manner. That is because, once

122
ŒÆ b s a ªØŒ Æ Æ, a b NØŒ Æ Æ K IæØŁfiH  ØØ, P c
Iæø fi · a b ¼ Æ, –æ K d a  a a NØŒ Æ Æ, ¼ØæÆ. Øe ¼åæØ H
NØŒø  ø Ie H ªØŒø  ø ŒÆ Ø Æ Æ挺 › —º ø Æ ŁÆØ,
ŒÆ ØÆØ b Øa H Øa  ı ØÆØæ F Æ ÆE N  Ø E ØÆç æÆE· a b ¼Øæ
çÅ Ø KA, c ªaæ i ª ŁÆØ  ø KØ Å. ŒÆ Ø ø b s N a NØŒ Æ Æ
IªŒÅ ØÆØæ F Æ Øa º Ł ı NÆØ, IØ ø b N a ªØŒ Æ Æ IªŒÅ ıÆØæE
e ºBŁ  N " (ed. Busse 1887).
123
So Szlezák 1972. Certainly the associations of the text, including Andronicus
and Eudorus, belong to the first century, and there is no trace of the dominant
later discussions that we encounter in (say) Sosigenes in the second century (cf.
Dexippus, 7,1 and following). Hippolytus (170–236 ce) seems to allude to the text
in his Refutations (6.24), and Iamblichus uses it extensively in commenting on the
Categories.
 Andronicus of Rhodes 71
they grasp reality and witness real beings, they apply the fitting mean-
ings absolutely.124 (31,6–32,23 Thesleff )
This passage seems to express a similar common position, including
the emphasis on transcending vagueness and ignorance, and on
distinguishing accidental from essential predication, with additional
Pythagorean undertones.125 I suggest that Andronicus developed this
position in the first place, stressing the value of the Categories as a
crucial initial tool for undertaking the kind of division necessary for
generating formal definitions—shifting from mere æ º łØ based
on vague sensible particulars to properly articulated  ØÆØ—and
defended its value on this basis.

(e) Other testimonies on the Categories

I have focused here on testimonies to Andronicus’ interest in the
general function and purpose of the Categories, and its goal or target

124
The text runs as follows (excerpted from Szlezák 1972): Kd b a Z Æ ØåH
ºª ÆØ· a b ªæ K Ø  ŒÆ, a b ııæå  Æ j ıÅŒ Æ  Ø· ŒÆd
H ııÆæå ø j ıÅŒ ø  Ø e  K Ø  Ø Å, b æ  ø
å  ŒÆd a ¼ººÆ  Ø B ŒÆd › ø· [ . . . ] (30) e b s  ÅÆ F Æ e
ºª ı, a ııæå  Æ ÆP fiH Æ  ŒÆd › Æ ŒÆd H åØ æe ¼ººÅºÆ
ıºŒ  ŒÆd ØÆØæ Æ  ºø ÆØ. KŒ b  ø çÆæ K Ø, ‹ Ø
¼Łæø  ŒÆ [p. 32] K Ø ŒÆd ÆŁc B Z ø KØ Å ŒÆd ªª  æe e
ŒÆ ØE c H Z ø –Æ Æ Iº ŁØÆ· P ªaæ ¼   çı   ŒÆd ıªªB F
ºª ı F Æ e r å c ŒÆ ÆææŁØ Ø, Ø´ w æ ŒÆd a ŁEÆ H æƪ ø ŒÆd a
IŁæØÆ ŒÆd a ªª  Æ ŒÆd a ªØÆ ŒÆd a ªÅ Æ æÅ ÆØ ŒÆd
ØÆ ÅÆ ÆØ·  Æ ªaæ j K  Ø j Øa  ø j PŒ ¼ı  ø åØ c
ŁøæÆ· j ªaæ æd P Æ › ºª  j æd H ııÆæå ø ÆP fi B ŒÆd X Ø æd
Æ  j I ø ı ŒÆd H I ø ø X Ø F K  Ø Å Ø j F K ªŁØ ŒÆd E
B.
(10) ! ¯ Ø Å b s A Æ KŒ æÆ ø Iæå Å [ H Nø Øåø] H
Iæø ª ÆØ [Iæåc] ªø ØŒ · o ø b c ÆØ åØ, hNi ŒÆd Ø´ Oºªø
ªøæÇØ.  b H Z ø KØ Å ŒÆd  ºf Aºº · Iæø ªaæ   Z Æ ŒÆd a
ªÆ ŒÆd a ªÅ Æ ªø ØŒ · F NŒ ø· › b ªaæ ¼Łæø  æe 
ÆF Æ ı Å, (15) æ  e r ÆØ ŒÆd e ŒÆºH r ÆØ· ŒÆd æe b e r ÆØ åÅ
KÅ  Æ , w åEæ IæåØ Œ , æe b e ŒÆºH r ÆØ KØ Å ŒÆd ºª 
æ Œ Æ ·  Ø b åÅ ŒÆd KØ Å A Æ  ƪ  Ø ŒÆd ‰æØ  · e b
Ø F  K IæØŁfiH. › b Æ IæØŁe › ŒÆ· NŒ ø ŒÆd a IŒæø æØÆ F
Æ  åØ ŒÆ ŒÆd a ØåEÆ (20) F Æ e ºª ı ŒÆ K d IºÅŁH. ƒ b
s ¼ºº Ø ¼Łæø Ø ŒÆ a ıÅŒe å ı Ø e ºª  ( P ªaæ N  åæH ÆØ), ƒ b
ç d ŒÆŁ’ Æ · I غÆÆ Ø ªaæ F Z  ŒÆd Łøæ F  a Z Æ KØçæ ı Ø
ŒÆ ’ ÆP H ÅÆ Æ ±æ Ç  Æ.
125
‘Archytas’, like Eudorus, privileges the absolute over the relative (see ch. 3).
72 Conclusions
in the later Neoplatonist vocabulary (æŁ Ø, Œ ).126 Androni-
cus’ other reflections on the text illustrate his attitude to Aristotle—in
general, he engages him more or less as an equal, without the sort of
reverence that characterizes even Boethus’ treatment (Moraux, Aris-
totelismus I: 99). For example, he famously replaced Aristotle’s cat-
egories of Where and When by Place and Time (Simplic. in Cat.
134,5, 342,23, 357,28). He added a fifth kind of quality to Aristotle’s
own list (263,19–22). He seems to have suggested—if this is not a
textual criticism but a philosophical point—that Aristotle could have
expressed himself more clearly at Cat. 6a37, as his extant definition of
the relative is circular (202,5).127 All of these innovations sparked
defences of Aristotle by Andronicus’ successors, especially Boethus.

IV. CONCLUSIONS: ANDRONICUS’

RELATIONSHIP TO THE ‘ANCIENT EXEGETES’
OF CATEGORIES

Based on the limited material at our disposal, there is no strong

evidence of Andronicus responding critically to alternative interpret-
ations of the Categories, such as those offered by Lucius (discussed in
ch. 4). The only alternative reading of the treatise that Andronicus is
clearly concerned to rebut is that which pegs it as an introduction to
the Topics, and so to plausible dialectic rather than scientific demon-
stration. Still, we might attempt to develop a tentative narrative of
the course of influence of Andronicus’ promotion of the Categories,
focusing on the ‘ancient’, or first-generation, readers of the text
(Simplicius in Cat. 159,33: f ƺÆØ f H ŒÆ Ū æØH KŪŠa).128
In this section, in anticipation of chs. 3–6, I sketch this narrative.
(Again, I would like to stress at the outset that I do not take

126
On which see for example Mansfeld 1994: 1–9, 20, Hoffmann 1987, Hadot
1991.
127
I attempt to set out some of Andronicus’ other contributions more generally in
Griffin 2009: ch. 3.
128
Namely, Boethus of Sidon and Ariston (Peripatetics), Eudorus of Alexandria
(who may be called an ‘Academic’, or Pythagorean; cf. Bonazzi 2013, 160–2), and
Athenodorus (a Stoic), to whom we might add the shadowy critic Lucius, who is
mentioned only by Simplicius and is not listed among ‘the ancients’, but seems to
belong there chronologically, since Boethus responds in some detail to his critique.
 Andronicus of Rhodes 73
Andronicus’ ideas about the text to be especially influential on his
contemporaries, with the exception of Boethus of Sidon; rather,
I argue that his simple promotion of the text—from relative obscurity
as a preface to the Topics to the fore of his catalogue—may have been
the primary motivator of wider interest in the treatise Categories.)

Peripatetics
Let us turn first to Andronicus’ Peripatetic near-contemporaries,
Boethus and Ariston. Since there is good evidence of Boethus of
Sidon either correcting or defending and expanding Andronicus’
interpretations of Aristotelian philosophy (Aspasius in Eth. 44,24),
and plentiful evidence of his response to Andronicus’ reading of the
Categories in particular, we can support the ancient tradition (e.g.
Philoponus in Cat. 5,19) that Boethus was influenced by Andronicus,
and perhaps in some formal sense his pupil. We can locate Boethus
after Andronicus—not necessarily chronologically, but simply in
terms of influence, suggesting that Andronicus’ publication, para-
phrase, and foregrounding of the Categories helped to attract the
kind of detailed interest that Boethus displays in the treatise. In
particular, as I argue below in the chapter on Boethus, Boethus adopts
Andronicus’ vision of the Categories as a useful introduction to
philosophy for beginners.
There is very little evidence to tell us much about Ariston (likely of
Alexandria), a newly-minted Aristotelian129 with whom Andronicus

129
On Ariston, see for example I. Mariotti (1966) and Moraux I, 181–93, with
Caujolle-Zaslawsky and Richard Goulet, ‘Ariston d’Alexandrie’, in Goulet (1989) and
Gottschalk (1987), 1110 and 1120–1. For his ‘conversion’ to Aristotelianism, see now
Blank 2007: 92 and Chiaradonna 2013: 39–40, with Moraux 1973: 225–6 and Donini
1977: 247–8. Strabo (Geog. 17, 1, 5) is a chief witness for the life and chronology of
Ariston. He must have lived during or before Strabo’s own life. Fraser (489) argues
that Strabo was unlikely to have known Ariston during his own sojourn to Alexandria
in and after 24, as the latter was a pupil of Antiochus (and his brother), and there is a
lapse of forty years between the probable death of Antiochus in 69 and Strabo’s
residence in Egypt. The information about Ariston’s relationship to Antiochus
depends on Philodemus, Index Acad. Herc. col. 35. The Philodeman Index has Ariston
move from the Academy to the Peripatos with Cratippus. It is not demonstrable that
the Academic–Peripatetic Ariston mentioned in the papyrus alongside Cratippus is
the same Ariston of Alexandria mentioned by Strabo. It was assumed, e.g. by Moraux
(1973: 57), although Moraux also recognizes the uncertainty of the identification (182;
cf. Tarán 499). If they are the same, certainly this is some help; Cratippus’ activity was
contemporary with Cicero, who mentions him as the leading Peripatetic in Athens (cf.
74 Conclusions
shared a bid to revise the definition of the Relative at Cat. 8a31 to
avoid circularity; Andronicus, Ariston, and Boethus are all credited
with this argument,130 although perhaps the fact that Porphyry men-
tions only Andronicus (in Cat. 125,20) is mildly suggestive in his
favour as its originator. This evidence, I think, only points to the
possibility that Andronicus was one of several contemporary Peripat-
etics, like Ariston and Cratippus, who grew increasingly interested in
the Aristotelian esoteric texts and in working out and correcting the
ideas contained in the Categories in particular—though they were not
yet driven by the commentarial, line-by-line approach that Boethus
would subsequently adopt.

Platonists
Before turning to Eudorus, let us consider Lucius, a very influential
critic of the Categories (discussed in more detail in ch. 4). Lucius, who
has Platonist sympathies, forcefully objects to the Categories on the
grounds that it professes (but fails) to offer a compelling account of
being, since it omits any discussion of intelligible being at all: in other
words, it’s a failed shot at a system of first philosophy, or ontology.
While ‘Lucius and Nicostratus’ refer to Andronicus in the course of
their objections to the Categories (ap. Simplic. in Cat. 134,4, in the set
of aporiai at 127,11–140,30)131 and they rely on some of Andronicus’
interpretations, we have no evidence that Andronicus was in turn
aware of the interpretations and challenges to the text offered by
Lucius. Boethus of Sidon, however, certainly was; he responds to their
objections systematically (see ch. 4). We might, then, suggest this

De Off. 3.2, De Div. 1.3, etc.—although as Gottschalk and others stress, this need not
imply that he was ‘scholarch’); thus the close attachment of Cratippus with Ariston
could also weight Ariston’s activity before the later forties bce.
130
Simplicius (in Cat. 201,34) states that Ariston and Boethus changed ti to
heteron in the definition of the Relative at Cat. 8a31 to avoid circularity; Andronicus
appears to have done the same. Simplicius subsequently states o ø b ŒÆd
æØŒ , although it is not clear who follows whom: there is perhaps a superficial
reason to favour Andronicus, who is mentioned alone at Porphyry, in Cat. 125,20 ff.
I think there is no immediate reason to doubt that Andronicus here originated the
interest in the discussion of the relative from an Aristotelian standpoint, and shared
this interest with Ariston. Tarán 1981: 742, suggests Stoic influence on Ariston;
Gottschalk 1987: 1106 n. 138, points to Aristotelian sources instead.
131
This, however, does not prove very much chronologically, since Simplicius
often reports Lucius, Nicostratus, and Plotinus as if they presented a united front.
 Andronicus of Rhodes 75
rough sequence of events: Andronicus’ new emphasis drew serious
attention to the Categories outside the circle of scholars like himself,
Ariston, and perhaps bibliophiles like Tyrannio; Lucius represents a
critical response from a Platonist standpoint (which would imply a
relatively early date for Lucius; on this, see ch. 4); Boethus, belonging
to the next generation, replied to Lucius’ objections while also devel-
oping a more subtle reading of the text’s metaphysical implications
and amending Andronicus’ views accordingly.
But why did Lucius respond so vigorously against an interpretation
of the Categories that simply—with Andronicus—made it into a
useful, introductory handbook for teasing out our ‘preconceptions’
and developing demonstrative technique, a foundation for logic?
What threat was there in that? For an answer, we must turn to
Eudorus, a near contemporary of Andronicus who shares his analysis
of the Categories under the broader subdivision of the absolute and
the relative, and the conviction that its study can lead to demonstra-
tive knowledge (these affinities are discussed further in the following
chapter). Eudorus, followed in several important respects by Pseudo-
Archytas, adopted these insights in a Pythagorean and Platonist
framework, arguing that the Categories offers an accurate map of
the structure of reality, both intelligible and sensible, once it is lightly
amended (for example, to position quality before quantity, represent-
ing the Monad before the Dyad). It appears to be this kind of
interpretation, not Andronicus’, that sparked objections from the
school represented by Lucius. It is difficult to judge the direction of
influence here, although Andronicus’ motivations for treating the
Categories as he did display no evidence of Eudorus’ specifically
Pythagorean innovations or concerns. We might then suppose that
they evolved independently and that the attention that Andronicus
drew to the treatise helped to attract interest in Alexandria, although
others introduced a very different, metaphysical and ontological set of
concerns, which then in turn attracted critical engagement from other
Platonists. Eudorus may well have developed his interests in the
Categories entirely independently from Andronicus, and for different
reasons (see ch. 3).
So perhaps we might update our suggested narrative as follows:
Andronicus’ methodological and pedagogical concerns motivated
his foregrounding of the Categories in his catalogue, which in turn
helped to draw attention to the text; Eudorus represents a growing
Alexandrian school of interest in the Categories as a metaphysical or
76 Conclusions
ontological treatise offering a map of reality, which in turn attracted
other Platonist interest and critics such as Lucius; Boethus responded
to them in turn, arguing in his turn that the Categories was only
intended to offer a map of sensible reality, and metaphysicians would
be better off exploring its relationship to the central books of the
Metaphysics. Boethus’ view, of course, would prevail through Por-
phyry, with Plotinus adopting the Lucian and Nicostratan critique in
the meanwhile.132

Stoics
This narrative omits just one of the ‘ancient’ readers of in Cat. 159,22,
namely the Stoic Athenodorus. There is no clear evidence, I think,
that Athenodorus’ interest in the Categories was stimulated by An-
dronicus; in fact, there is some prima facie reason to doubt that, since
Athenodorus’ worry about the treatise is that it does a poor job of
analysing natural language into its component parts. That is to say, it
is a bad textbook of grammar, in the sense that the tenfold division
seems to be the wrong number (cf. Simplicius in Cat. 62,24–6) to
capture everything from conjunctions to prepositions, nouns to verbs
(for this argument, see ch. 5). This would be a very odd and blinkered
response to the Andronican interpretation that this treatise is about
meaningful predication (katēgoria), especially the clear and precise
distinction of absolute from relative predication, functioning as an aid
to definition and demonstration, and that the book provides an on-
ramp to philosophical demonstration through the articulation of our
innate æ º łØ.
Perhaps, instead, we might locate Athenodorus as an earlier, Stoic
respondent to Aristotle’s theory of ten kinds of predication, under-
stood as useful for rhetoric or grammar, compatible with its earlier
reading under the rubric of the Before-the-Topics or the rhetorical
interpretation found, for instance, in Quintilian Inst. 3.23–34. (This
might make it chronologically less likely, though not impossible, that
the ‘Athenodorus’ mentioned by Simplicius is Athenodorus Calvus,
the tutor of Augustus, but there are other, earlier alternatives for that
identification.)133 There is a little evidence for Boethus of Sidon

132
See Chiaradonna 2002: 263–7.
133
I argue this in more detail below and especially in Griffin 2013b.
 Andronicus of Rhodes 77
responding to Athenodorus’ concerns, but not very much: Boethus
responds, here elaborating and adding precision to Andronicus’ view,
that the treatise is not trying to analyse verbal expressions as such
(ºØ ŒÆŁe ºØ) but instead simple verbal expressions qua sig-
nificant of being—that is, it is a work of semantics. This influential
Boethan interpretation would prevail through Porphyry. (There is
also some evidence, as I suggest in ch. 4, that later Stoics like Cornutus
agreed that the Categories belonged to the primary part of dialectic
in the Stoic sense, dealing with ºŒ  and not with grammar as
such, although there are also passages, such as 359,1, that speak
against this.)
The following chapters aim to develop and elaborate this narrative.
 3

‘Pythagorean’ Support: Eudorus and

Pseudo-Archytas

There are strong traces of interest in Aristotle’s ten categories—and

the treatise Categories—in Pythagorean circles roughly contemporary
with Andronicus, beginning with Eudorus of Alexandria. This inter-
est seems broadly similar in temperament to Andronicus’, treating
the ten categories as a crucial aid in transcending the vagueness of
particulars and arriving at the more precise knowledge of intelligibles,
but with a new, Pythagorean emphasis on the symbolism of the
number ten and the recognition of the distinction of the intelligible
world from the sensible (compare passages cited above, especially
Pseudo-Archytas, —æd H ŒÆŁºı ºªø XØ H ŒÆŪæØH,
31,6–32,23). I will briefly trace this interest here, which has been
studied in full detail recently, especially by Mauro Bonazzi (2007,
2012, 2013).

I. EUDORUS OF ALEXANDRIA (FL. c.50 bce?) 1

Life and History

We have several hints regarding the life and chronology of Eudorus,
suggesting that he was active around the middle of the first century
bce. The chief evidence is Strabo (64 bce–19 ce), who mentions

1
The fragments and testimonia for Eudorus of Alexandria have been collected by
C. Mazzarelli, ‘Raccolta e interpretazione delle testimonianze e dei frammenti del
medioplatonico Eudoro di Alessandria’, Rivista di filosofia neo-scolastica 77 (1985),
197–209 and 535–55; see also A. N. Zoubos, ‘¯høæ › º Ææ ’, Athena 62
 ‘Pythagorean’ Support 79
(Geog. 17.1.5) that Eudorus and Ariston both wrote similar books On
the Nile, and Eudorus accused Ariston of plagiarism. We can infer
that Eudorus and Ariston were active before or during the lifetime of
Strabo himself. As Strabo declares that he cannot tell which book
really came first (K @ ø oæØ Ø ¼!), we may also infer that
Eudorus was relatively well known in that time.2 With respect to his
school affiliation (on which see also Bonazzi 2013), Eudorus appears
to have been regarded as an ‘Academic’ by his near-contemporaries,
and represents his own sympathies as Pythagorean.3

Works
(1) Eudorus is said (by Arius Didymus ap. Stobaeum, loc. cit.) to have
written a ˜ØÆ æØ F ŒÆa çغç Æ ºªı. According to Arius,
this text was arranged æºÅ ÆØŒH, by topic or problem. Arius
preserves a portion of the ethical section of the work,4 and praises it as
a work well worth purchasing (غ  I ØŒÅ). This is the only

(1958), 194–203 (with modern Greek commentary). Literature and discussion of

Eudorus during the past century is collected in Goulet (2000:292–3). Tarrant (2008)
has specifically explored Eudorus’ polemical posture with regard to the Cat. On
Eudorus see in particular Dillon (1977, 1996, and 2000) and recently Chiaradonna
(2009 and 2013), Bonazzi (2007), Fraser (1972), Dörrie (1944), Tarrant (1985 and
2008), Rist (1986), and Napolitano (1985).
2
Unlike Ariston, Eudorus is not mentioned by Cicero. This could suggest that he
was less conspicuous in the years before 43 bce, but this is inconclusive: for example,
as Dillon points out (1977/1996: 115), Cicero may not have been in contact with
Alexandria’s latest developments of philosophy in the years leading up to 43. It is
difficult to guess when Eudorus produced the book that brought him into conflict with
Ariston, or for that matter when he produced major publications (see below) that
could have drawn him to the attention of a figure like Cicero. We may, however,
hazard one speculation about his relationship to Ariston; it seems unlikely that
Eudorus was a relative unknown, a newcomer to the Alexandrian scene, when he
accused Ariston of plagiarism; it seems from Strabo’s report that both persons were
fairly well known in Alexandria.
3
Arius Didymus, who was also active in the first century bce (if he is to be
identified with the confidant of Augustus: on Arius, see Runia in Der Neue Pauly
and Goulet, DPhA s.v.) calls Eudorus an ‘Academic’ ( ŒÆÅ ØŒF çغçı, ap.
Stobaeum, Ecl. 2.42,7 ff. Wachsmuth-Hense), as does Simplicius (› ŒÆÅ ÆœŒe, in
Cat. 187,10). Eudorus himself also appears to have traced his views to those of the
Pythagoreans (see for example Simplic. in Phys. 181,10), and to have sought to
reconcile representatives of the Old Academy within a shared Platonic tradition
(Plutarch).
4
Meineke (1859) initially attributed the ethical passage in Stobaeus to Arius. On
Arius and his identification with the Stoic contact of Augustus, see David Runia in Der
80 Eudorus of Alexandria
work of Eudorus’, as it happens, that we know by title. Several further
works and commentaries are inferred from fragments and testimonia,
but I take it that, in reality, any of the following testimonies may
equally well derive from essays in the ˜ØÆ æØ. (2) Eudorus also
commented on Plato’s Timaeus, as we learn from Plutarch On the
Generation of the Soul (see 1013B5, 1019E6, 1020C3; several passages
are discussed below). (3) Eudorus wrote about the Metaphysics of
Aristotle, as we learn in Alexander of Aphrodisias (in Met. 59,7,
where he proposes an emendation, perhaps to show that in Plato’s
more esoteric or secret doctrine, God is the first cause of both the
Forms and Matter). (4) Eudorus wrote about the Categories of Aris-
totle, as discussed below; again, however, it is unclear whether this
represents a separate commentary. (5) A papyrus about optics (POxy
1609) mentions a Timaeus commentary by the same author, and
there is a chance that this could refer to Eudorus (see Dillon 2000).5
(6) The book On the Nile mentioned above, concerning the cause of
the river’s flooding, is presumably another independent work by
Eudorus, and (7) he produced a work of commentary on astronomy,
perhaps on Aratus (see Dörrie and Baltes, 2002: 290).

Philosophy
Eudorus has often been labelled an exemplary ‘Neo-Pythagorean’
(e.g. Fraser 1972: 493), and he was labelled an ‘Academic’ in antiquity
(see Bonazzi 2013: 161–2). Indeed, Dörrie suggested that he origin-
ated the Pythagorizing ‘rebirth’ of Platonism in the first century
(32–4, referencing Simplic. in Phys. 181). With Dillon (1996/1977;
cf. Dillon 2000), we might see him as motivated to return to the
doctrines of the ‘old’ Pythagoreans, Archytas and Philolaus, and the
‘old’ Academics, Speusippus and Xenocrates, thus sharing a common
motivation with Antiochus of Ascalon. (This is one respect among
many in which Eudorus’ sympathies are remarkably close to Andro-
nicus’, as we shall also find below: the latter’s sympathetic treatment
of Xenocrates’ psychology at Themistius in An. 31,1–32,34, and

Neue Pauly, Brad Inwood s. v. ‘Arius Didymus’ in Goulet (1989), and Moraux
(1973:259–450).
5
Burnyeat (2005) has recently argued that the original Archytas of Tarentum
founded the Greek science of optics, beginning a tradition in which Eudorus may
have worked.
 ‘Pythagorean’ Support 81
affiliation with Xenocrates at Simplic. in Cat. 63,22–8, represent a
similarly favourable attitude toward the ‘Old Academy’).
Eudorus marks an early example of the strong preference to ‘har-
monize’ canonical philosophers that thrived in ancient philosophy
from the first century onward.6 Plutarch provides an example of
Eudorus’ approach to the Old Academy. Here Eudorus is represented
as seeking to harmonize the definitions of the soul promoted by
Xenocrates and Crantor in order to produce a unified account of
the soul.7 Eudorus’ own view is not entirely clear, but what Plutarch
does make clear is that Eudorus seeks to show that Xenocrates and
Crantor were not at odds, and that neither was altogether mistaken.
We can be reasonably confident that he thought the Academics were
on the right track, and he wanted to show that they were not at odds.
An instance of Eudorus’ exegetical approach to earlier ‘Pythagor-
eans’ can be found at Simplicius in Phys. 181,7–30, a verbatim
quotation in which Eudorus ascribes to ƒ —ıŁÆªæØØ the view
(which he himself endorses) that there exist a ‘Monad’ and a ‘Dyad’
under which all the Opposites are ranked. The ‘Opposites’ refer here
to the Pythagorean opposites such as (on the one hand) limit, odd,
male, light, and (on the other) unlimited, even, female, dark. But
above the Monad and the Dyad reigns a supreme One, which may be
described as the ‘transcendent God’ (æø Ł), and is the cause
both of Matter (oºÅ) and of all created things.8

6
On this tendency in general, see for instance Boys-Stones (2001) and
Karamanolis (2006).
7
See also Bonazzi (2013: 164–71). Dörrie, Hermes 79 (1944), 27–8 argues that
Eudorus reconciled Xenocrates’ symbolic exegesis with Crantor’s more ‘literal’ exe-
gesis. (Xenocrates argued that the soul was a ‘number moved by itself ’, while Crantor
described it as a mixture of noētē ousia with ‘that which forms impressions of
perceptible objects by means of opinion’.)
8
Eudorus ascribes the view that the One causes Matter, as it seems, to Plato
himself as well. Thus we learn from a controversial passage of Alexander in Met.
(59,7), in which Eudorus—preserved for Alexander via Aspasius—is said to have
amended the text at Aristotle, Met. 1.6, 988a7–12 to indicate that—for NØ, i.e.
‘those who know’ the true doctrine of Plato—the One is the cause of all things, even of
Matter, being superior to the Forms (discussing Met. 1.6, 988a7–12). Moraux has
discussed the text in detail in ‘Eine Korrektur des Mittelplatonikers Eudoros zum Text
der Metaphysik des Aristoteles’, Beiträge zur alten Geschichte und deren Nachleben,
Berlin, 1969, vol. 2, 592–04. Bonitz had proposed to amend the variant quotation at
59,1–2 to print YØ for NØ (cf. Met. 988a10–11), but YØ does not seem
compatible with what Alexander says next. Moraux would take ŒÆN B fi oºfiÅ as outside
the variant reading, to read something like: ‘it is clear that [Plato] used only two
causes, the essence (for the Forms are causes of the essence for the other things, [and],
82 Eudorus of Alexandria
Treatment of the Categories
Eudorus applies these views to his reading of the Categories. We have
some direct evidence, and Philo of Alexandria’s use of the Categories
provides an indirect source.9

1. The categories fall under the absolute and the relative

First, at Simplicius, in Cat. 174,14, Eudorus suggests that the ten
categories can be grouped into two major classes, the Absolute (ŒÆŁ’
Æe) and the Relative (æ Ø)10—adopting a similar view to that of
Andronicus (see ch. 2).11 These two ‘Platonic categories’ can perhaps
be traced to Xenocrates (fr. 12 Heinze),12 who is also mentioned in
concert with Andronicus by Simplicius, raising the possibility that
Andronicus joined in a wider reflection on Old Academic dialectical

for those who know, the One), and matter’. If that reading is right, Eudorus did not
attempt to interpret Aristotle’s account of Plato according to a more ‘Middle Platonic’
view; Moraux 503). But Dillon suggests (1977:128 n. 1) that Eudorus intentionally
makes the One the cause of hulē as well as the eidē, noting that this ‘Eudoran’ monism
appeals to Philo and appears later in the Chaldaean Oracles (e.g. fr. 34 Des Places and
Psellus Hypot. 27).
9
On Philo, see Runia in Der Neue Pauly [I 12], with extensive bibliography. On
his use of Peripatetic material, see Gottschalk, 1141 and 1145–6. Philo, a key repre-
sentative of Hellenic Judaism, is not mentioned by the later commentators in associ-
ation with the Categories, and his voluminous corpus does not include any systematic
commentary on the Aristotelian text. For instance, he is not one of the ‘ancients’
named by Simplicius (in Cat. 159,32–3). However, it is worth mentioning at least
briefly that Philo makes some use of the theory of categories, within a Pythagoreaniz-
ing framework (The index in Borgen et al. 2000 is very useful for identifying refer-
ences; Dillon (1977: 178–82) usefully summarizes key passages.) In one passage, Philo
claims that there is ‘nothing that does not participate’ in the ten categories (cf. Dec. 30
and following). In particular, like Eudorus, Philo uses the order Ousia–Quality–
Quantity for the categories (cf. Dec. 30 and following); in the same text, he gives the
name topos and chronos for Place and Time—not retaining ‘Where’ and ‘When’. But
Philo locates Time and Place at the end of the list—as does ‘Archytas’, whilst Eudorus
places them immediately after Quantity. This passage suggests that Philo’s source for
the Categories is working in the same tradition as Eudorus and Philo.
10
On Eudorus’ comments here, cf. Szlezák (1972:130) and Sharples (2008:282).
11
The twofold division is also found in the Divisiones Aristoteleae (DL 3.108 and
following = Cod. Marc. 97, p. 39 Mutschmann). Tarán, in reviewing Moraux I,
suggests that Andronicus is influenced by Stoics as well as Xenocrates; Gottschalk
(1990:72) disagrees.
12
Dillon (1977:133) points out that the view may be traced to Plato himself (on the
testimony of Hermodorus ap. Simplic. in Phys. 248,2 Diels); in the Sophist 255C, Plato
distinguishes auta kath’ hauta from pros alla.
 ‘Pythagorean’ Support 83
thinking, which might even have been spurred in the first place by
Alexandrian Academics like Eudorus. For all we know, the ten-
categorial scheme that we associate with Aristotle’s Categories may
also have been defended by Xenocrates, with important differences.
I would stress here the distinction between the doctrine that the ten
categories helpfully subdivide the absolute and the relative, and
therefore accurately carve reality, which Eudorus and Andronicus
share, and Aristotle’s discussion of the ten categories in the treatise
Categories, which Eudorus may have criticized more heavily than
Andronicus.13 It is important to notice, also, that Eudorus’ move to
bring the ten categories under the Academic scheme of two is not
necessarily a piece of school polemics. As Chiaradonna succinctly
observes, ‘maintaining the Platonic–Academic twofold division of
categories along with Aristotle's list cannot in any way be seen as a
sign of anti-Aristotelian allegiance’ (2013: 47–50): there is no evi-
dence here that Eudorus was treating the Categories from a hostile
standpoint.
The later Pythagoreans followed (a view that they understood to
be) orthodox ancient Pythagoreanism when they mapped the two
Platonic categories, absolute and relative, to the two columns of the
Table of Contraries—ten opposed pairs organized in a table, ŒÆa
ıØå Æ (cf. Aristotle, Metaphysics 986a22–6):

Absolute (ŒÆŁ Æ) Relative (æ Ø)

Limit (æÆ) Unlimited (¼Øæ)

Odd (æØ) Even (¼æØ)
One () Plurality (ºBŁ)
Right ( Øe) Left (IæØæ)
Male (¼ææ) Female (ŁBºı)
Resting (Mæ F) Moving (ŒØ )
Straight (PŁ ) Curved (ŒÆ  º)
Light (çH) Dark (Œ)
Good (IªÆŁ) Bad (ŒÆŒ)
Square (æªø) Oblong (æ ÅŒ)

13
I am very grateful for correspondence with George Boys-Stones that has helped
me to clarify this distinction.
84 Eudorus of Alexandria
For instance, absolute and relative correspond respectively to odd and
even, and limit and unlimited.14 We also know from Simplic. in Phys.
181,7–30 that Eudorus maintained that this division of the opposites
was the orthodox Pythagorean view, which he appears to endorse.
There is no reason, I think, to suppose that this table is designed to
compete in any way with the Aristotelian ten categories, although it
does suggest a common interest in the number ten—in this case, the
perfect number of the Pythagorean tetraktys—as capturing a com-
prehensive model of reality. The table simply illustrates the two
Academic categories using terms drawn from a Pythagorean lineage,
to emphasize the antiquity and authenticity of the Academic scheme
(for common moves of this kind beginning in the first century bce
and ce, see for example Boys-Stones 2001: 101); the same Academic
categories can also be used to explain and illustrate the ten categories
that we associate with Aristotle without conflict or competition. (One
possible, speculative connection between the two is the stress laid on
division through contraries by Aristotle, for example at An. Post. 2.13,
97a14–23; I have suggested above that the account of division and
definition in these chapters of An. Post. 2 was at any rate influential
for Andronicus. If we are to divide by differences, we ought to use
‘opposites with nothing between them’: for example, ‘line’ might be
divided into ‘straight’ and ‘curved’. Possibly having a list of funda-
mental contraries could be seen as useful for this project).
Eudorus also (Simplic. in Cat. 174,14–175,11) criticizes Aristotle
for treating only P Æ as ‘absolute’, and for examining the remaining
categories solely in ‘relative’ terms. Referring back to our discussion
of Andronicus in ch. 2, this might imply that Eudorus read Aristotle
roughly as Andronicus did, that is, as treating the non-substantial
genera in the Categories as all relative (æ Ø) in either a broad sense
or a strict sense. (Perhaps Eudorus was responding to Andronicus’
interpretation of Aristotle, or perhaps both were relying on a com-
mon source, or again perhaps Andronicus was influenced by
Eudorus’ critique). Nevertheless, as we will find below, Eudorus
clearly maintained the outline of the ten-categorial scheme that we
associate with Aristotle’s Categories, while arguing that the non-
substance categories, with the exception of relation strictly speaking
(which he, like Andronicus, treated as an appendage or offshoot of

14
See for example ‘Callicratidas’ De Dom. Fel. 103,12–13 Thesleff.
 ‘Pythagorean’ Support 85
the others: 174,18, 174,25), should be understood in their own right,
as predicable per se.
What is the point at issue between Andronicus and Eudorus here?
Andronicus may have presented Aristotle’s scheme as a model
in which one could only predicate ‘pale’ or ‘wise’ per accidens or (in
Andronicus’ ‘broad’ sense) as relative to the substantial subject in
which that qualitative feature inheres—as inherent, say, in Socrates.
When stressing how per se predication and per accidens predication
ought to be distinguished, Andronicus seems to have had in mind
cases where, on the one hand, substances are predicated per se of
substances (‘animal’ of ‘human’, ‘human’ of ‘Socrates’), or, on the
other hand, qualities or other non-substances are predicated per
accidens of substances (‘coloured’ of ‘body’, ‘wise’ of ‘Socrates’ soul’).
Clearly distinguishing these two kinds of predication is where Andro-
nicus’ focus fell, perhaps in order to play up the explanatory merits of
Aristotelian essentialism against the Stoic view. This Stoic position
was criticized by later Peripatetics, possibly following a line of argu-
ment inaugurated by Andronicus, for failing to distinguish essential
features from accidental features.
But it seems, based on the limited evidence available, that Andro-
nicus did not spend as much time in explaining cases of per se
predication in the non-substance categories (for example, ‘virtue’ of
‘courage’, or ‘colour’ of ‘white’, or for that matter self-predication,
such as ‘beautiful’ of ‘beauty’), and this omission may have been
perceived by Eudorus as the suggestion that Andronicus simply failed
to account for such cases. Aside from presenting problems for a
logical system, that would present a doctrinal stumbling-block for
Eudorus as an Academic, for whom qualities, as homonymous with
Platonic forms, ought to be self-predicating: wisdom (ç Æ) is, after
all, wise (ç), and wise par excellence, and beauty (ŒÆºº) is
beautiful par excellence.15 Eudorus, then, may have found himself
arguing more like Simplicius, when Simplicius rebuts as follows
Andronicus’ position that the non-substance categories are all some-
how relatives: ‘if we were . . . to refer an accident back to its subject . . .
taken in an absolute sense it is its own kind of nature’ (N s oø ŒÆd
e ı ÅŒe IÆçæØ  æe e Œ , Pb ŁÆı Æ,

15
For which see for example Phaed. 100C, Parm. 128E–129B, Symp. 211B.
86 Eudorus of Alexandria
Kd ÆPe ŒÆŁ’ Æe ç Ø N Æ K , 63,33–64,1).16 As the argument
against Andronicus continues, just matter (oºÅ), as the substrate itself,
should be considered properly relative (64,2)—that is, perhaps in the
strict sense, as properly having its existence only in its relationship to
other things, namely, the structures of quality, quantity, and so forth
that inform it. (And even matter, treated as a principle rather than as a
substrate, can be discussed per se: 64,3). Thus we should really under-
stand all the non-substance categories except relative (in the strict
sense) as predicable per se, which will suit the Academic reader better
than the Andronican interpretation.
This argument against Andronicus (for which Simplicius indicates
that he is following Porphyry and Iamblichus, 64,3–4) might even be
traced back to Eudorus, with the important caveat that for Eudorus,
this argument counts against Aristotle’s version of the ten-categorial
scheme, since he does make the non-substance categories relatives.
For the later Platonists, the trouble is with an (Andronican) misin-
terpretation of Aristotle, grounded in inspiration from Xenocrates,
and not with Aristotle himself. This presents us with the simple
possible interpretation that Eudorus read Andronicus’ interpretation
of Aristotle’s Categories, recognized the value of the ten-categorial
scheme offered by Aristotle, but ‘corrected’ Aristotle to make the
doctrine ‘safe for the Academy’ (whatever ‘Academy’ might imply
doctrinally here. I have suggested above that it might imply, at least,
the requirement that qualities and the like could be self-predicating).
Others, perhaps responding to Eudorus or acting on their own, felt
that only a correction of Andronicus’ interpretation was required. But
this would leave to one side the fact that Andronicus was apparently
already influenced by Academic thought, at any rate by the Old
Academy of Xenocrates, in bringing the scheme of ten categories
under the rubric of the absolute and accidental—he just did so a little
differently. Eudorus was obviously steeped in Xenocrates (I also
discuss this relationship a little further below), and Andronicus, too,
had read and occasionally defended Xenocrates (see ch. 2). There is a
fair possibility, then, that both Eudorus and Andronicus were directly
or indirectly familiar with Old Academic treatises interpreting and
revising the ten-categorial scheme shared by Aristotle’s Categories
(and by Aristotle’s use of the ten categories throughout his thought).

16
See Dexippus 31,28ff. Of course, it is also perfectly legitimate to say that ‘pale
exists’ (so Simplic. in Cat. 42,21–2) or ‘wisdom exists’.
 ‘Pythagorean’ Support 87
Whether or not we entertain this possibility, it is also, of course,
perfectly reasonable to suppose that Eudorus’ and Andronicus’ inter-
est in the scheme of ten categories was independent; or again, it is also
possible that at least one of the two was aware of the work of the
other.17
But if indeed Andronicus’ interest in the treatise Categories was
motivated by a recognition of the value of distinguishing absolute
from relative predication, and that recognition was in turn inspired
by interpretations and revisions of the ten-categorial scheme in the
Old Academy, then we can draw the following interesting if specula-
tive conclusion: perhaps a resurgence of Academic thought in general
in the first century—as represented by Eudorus—can also be credited
with the first burst of interest in the treatise Categories.

2. The order of the Categories: evidence for

an ontological interpretation
At Simplicius, in Cat. 206,10–15, Eudorus focuses on the ordering of
the ten categories, placing quality after P Æ, then where and when,
which he relabels Place and Time (as did Andronicus):
Eudorus too says that the discussion of Quality is closely linked with
that of Substance, and that the discussion of Quantity belongs after that;
for Substance coexists with Quality and Quantity, and it is after these
that the categories of Time and Place are introduced: for all substance is
somewhere and at some time [viz. sensible substance] (tr. Fleet 2002,
lightly modified).18
Eudorus’ following explanation seems to be clear: after P Æ, the
remaining ten categories should be organized based on the closeness
of their relationship to P Æ. Quality should come second (a view
also taken by Lucius and Nicostratus) because P Æ cannot exist
without Quality; Quantity follows for the same reason; and because

17
If the story that Eudorus accused an Ariston of plagiarizing his book On the
Nile refers to the Ariston who is closely linked to Andronicus as one of the
‘ancient commentators’ on the Categories, that offers some circumstantial colour to
this suggestion.
18
ŒÆd ¯høæ b fiH æd B P Æ ºªø
fi e æd B ØÅ ºª ŒÆd a
F e æd F F ıÇFåŁÆ çÅØ· c ªaæ P Æ – Æ fiH ØfiH ŒÆd fiH
ııç ÆŁÆØ, a b ÆFÆ c åæØŒ  ŒÆd ØŒc ŒÆŪæ Æ
ÆæÆºÆ ŁÆØ· AÆ ªaæ P Æ F  r ÆØ ŒÆd , źØ c ÆNŁÅ
(ed. Kalbfleisch 1907).
88 Eudorus of Alexandria
all P Æ exists in Space and Time, these should be fourth and fifth.
We could infer from this that Eudorus takes the Categories to be
concerned with sensible P Æ; as a Pythagorean, we might reason, he
would hardly claim that intelligible P Æ is in space and time.
Indeed, Simplicius’ passage includes the gloss źØ c ÆNŁÅ
after P Æ (206,14–15) to make this clear, although it is not obvious
that the gloss is Eudorus’, or that Eudorus insisted that the Categories
addressed only sensible P Æ. He may, for example, have allowed for
an intelligible time and place in which intelligible substance might
subsist, as Iamblichus would later propose (see Dillon 1997). (I have
also suggested below some tentative reasons why Eudorus might have
allowed for the ten categories to apply to the intelligible world; on this
debate see Chiaradonna 2009b: 99.) By way of contrast, the author we
call ‘Pseudo-Archytas’, also discussed below, very clearly focused on
the sensible world alone as the scope of the Categories—at least its
account of the non-substance categories.19 (Interestingly, however,
‘Archytas’ placed time and place at the end of the list, like Philo of
Alexandria, perhaps representing the view that these are relatives;
see ch. 2.)
Why did Eudorus concern himself with the order of the ten cat-
egories and their respective degrees of distance from P Æ? He may
have anticipated Pythagoreans like Moderatus, who would metaphys-
ically rank Quantity as identical with the Dyad, with Quantity serving
as matter (oºÅ) for Form (see Moderatus and Porphyry in Simplicius
in Phys. 230,35–231,35); in such a framework, it might well be appro-
priate to prioritize Quality as identical with Form and the Monad.20
And like Moderatus, Eudorus postulates a ‘One’ transcending both
Monad and Dyad (cf. Alex. in Met. 59,7); he may have regarded this
One as the primary P Æ, the first of the categories. Then Eudorus
may have advanced a system of ten categories descending from a

19
So Szlezák (1972). ‘Archytas’ takes it as given that the non-substance categories
concern just the sensible world (cf. 22,31 and 30,21–3 and 31,5 Thesleff, Simplic.
in Cat. 378,1–3, also Dillon 134–5 for discussion: I am very grateful to Riccardo
Chiaradonna for directing me to the relevant passages here, although any misunder-
standings are my own). ‘Archytas’ agrees with Andronicus in separating the post-
praedicamenta, and with Andronicus and Ariston in changing Aristotle’s definition of
Relation (Simplic. 202,2). Archytas also differs from Eudorus on the ranking of
Relation, Space, and Time. He puts Relation fourth, after Quality and Quantity, and
Space and Time at the end of the list, in which he is followed by Philo.
20
See Dillon 1977/1996: 346–51.
 ‘Pythagorean’ Support 89
transcendent One or P Æ (corresponding to the first category), to a
Monad (quality) and Dyad (quantity), which in turn cascade through
the remaining categories as a table of Opposites falling somehow
under the Absolute and the Relative (cf. Simplic. in Cat. 174,14).21
That explanation would help to interpret both Eudorus’ concern for
the order of the ten Aristotelian categories and the careful distinction
that he draws between the One, on the one hand, and the Monad and
Dyad, on the other:
I affirm then that the followers of Pythagoras leave the One as principle
of all things, but in another way introduce the highest elements as two.
They call these two elements by many names. For one of them they
name ordered, definite, knowable, male, odd, right, light; the contrary of
that not-ordered, indefinite, unknowable, female, left, even, dark, so
that because in one way the One is principle, but in another the One and
the Indefinite Dyad are elements, both One[s] being alike principles,
and it is clear that the One as principle of all things is quite distinct from
the One opposite to the Dyad, which they also call Monad.22 (Eudorus
ap. Simplicium in Phys. 181,22–30, tr. Baltussen et al. 2012)
But Eudorus also maintained that Aristotle, in the Categories, had
delivered an incomplete account, one which treated the non-
substantial categories only in relative terms, and not in their own
right or absolutely (ŒÆŁ Æ):
Eudorus is critical, asking why, although the relative (æ Ø) is con-
trasted with the per se (ŒÆŁ’ Æe), Aristotle has discussed the relatives
and not the per se. It must be said that the [other] nine categories are
considered as being [within the realm of] what is per se. Therefore
establishing the per se throughout these nine categories he adds that of
the Relative as an appendage (ÆæÆçı Å) to these nine. For relative
is considered as being among the other nine with some ambivalence in

21
Such a Pythagorean account of the categories might even have found support in
Aristotle’s definition of P Æ as ‘One and the same in number’ and ‘receptive of the
Opposites’—which suggests both the descent of opposites from a Monad and Dyad,
and the ‘One’ that transcends both.
22
çÅ d  ı f æd e —ıŁÆªæÆ e b £ ø Iæåc IºØE, ŒÆ’
¼ºº b æ   a Iøø ØåEÆ ÆæتØ. ŒÆºE b a   ÆFÆ
ØåEÆ ººÆE æŪæ ÆØ· e b ªaæ ÆPH O ÇŁÆØ ƪ  ‰æØ 
ªøe ¼ææ æØe (25)  Øe çH, e b KÆ   ø fi ¼ÆŒ IæØ
¼ªø ŁBºı IæØæe ¼æØ Œ, u ‰ b Iæåc e , ‰ b ØåEÆ e
£ ŒÆd  IæØ ı, IæåÆd ¼ çø £ ZÆ ºØ. ŒÆd Bº ‹Ø ¼ºº  KØ £ 
Iæåc H ø, ¼ºº b £ e B fi ıØ IØŒ , n ŒÆd Æ ŒÆºFØ (ed.
Diels 1882).
90 Eudorus of Alexandria
the case of Substance, as he will say, but with none in the case of the
others; for example ‘state’ [is found] in Quality, ‘double’ in Quantity,
‘further and nearer’ in Where [ . . . ] in Substance there is ‘father-son’.
Therefore they think, that, since the category of the Relative is an
appendage to the others, it is supervenient, although it is prior and
considered in terms of its own differentia.23 (Simplic. in Cat. 174,14–26,
tr. Fleet 2002)
(If the report of Eudorus’ view continues throughout this passage, he
seems here to agree with Andronicus in calling the Relative an
‘appendage’ or ‘offshoot’ of the other nine categories, another
example of their correspondence: cf. Simplic. in Cat. 157,18–22. But
it was a commonplace that Aristotle himself regarded the relative as
an ‘appendage’: see Metaph. 14.1, 1088a23, EN 1.4, 1096a21–3.)
Perhaps Eudorus’ critique of Aristotle’s Categories should be taken
as implying that Eudorus (as a Platonist) insisted on the independent
and absolute reality of qualities (such as Justice and Beauty), quan-
tities (such as mathematical entities), and so on, whereas Aristotle
treated these merely as attributes of primary and secondary sub-
stances. Also, we should note that Philo of Alexandria, whose
treatment of the ten categories closely mirrors Eudorus and Pseudo-
Archytas,24 treats the characteristic of ‘being human’ as falling under
Quality, not the first category: possibly Eudorus had the same idea,
arguing on a Platonic line that (at least some of) Aristotle’s secondary
substances are really intelligible qualities or Forms, which can be
predicated per se. The same may have gone for quantities: certainly
some of the early readers of the Categories worried about whether

23
`NØAÆØ b › ¯høæ, Øa  IØØfiÅæÅ ı F ŒÆŁ’ Æe fiH æ Ø æd b
F æ Ø Ø ºŒÆØ › æغÅ, æd b F ŒÆŁ’ (15) Æe PŒØ. ŒÆd ÞÅ ‹Ø
K fiH ŒÆŁ’ Æe ƃ KÆ ŒÆŪæ ÆØ ŁøæFÆØ· e s ŒÆŁ’ Æe ºÆg ŒÆa a
KÆ ŒÆŪæ Æ c F æ Ø æ ŁÅØ ‰ ÆæÆçı Å ÆE KÆ. ŒÆd ªaæ K
ÆE KÆ ŁøæEÆØ, Kd b B P Æ I çغø ‰ KæE, Kd b H ¼ººø KÆæªH,
‰ Kd b F ØF   Ø, Kd b F F غØ, Kd b F F (20) KŒØÆ,
Kd , Kd b F b æ æ, æ, Kd b F ØE ŒÆd åØ ŒÆFØ,
 BØ, Kç’ z ŒÆd  KæªØÆ ŒÆd  EØ › ı Ø, Kd b F ŒEŁÆØ K ŒØÆØ,
ŒØÆØ, Kd b F åØ, Iç’ z e åØ Ææø ø ºªÆØ, ‰ Ø· ŒÆd Kd
B P Æ b Ææ, ıƒ. Øa ÆFÆ b ‰ ÆæÆçı Å ÆE ¼ººÆØ ŒÆŪæ ÆØ c
F (25) æ Ø KØØÅ  ÇıØ ŒÆ Ø æŪı Å sÆ ŒÆd ŒÆa
ØÆçæa NŒ Æ Łøæı Å (ed. Kalbfleisch 1907).
24
Philo adopts the Eudoran order P Æ–quality–quantity when he mentions the
categories (Dec. 30) and, like Eudorus and Andronicus, he replaces Where and When
with Time and Place (although like ‘Archytas’ and unlike Eudorus and Andronicus, he
places these categories last).
 ‘Pythagorean’ Support 91
Aristotle’s framework could embrace intelligible numbers and entities
like the ‘monad’ (Simplicius in Cat. 65,20–1). That interest seems to
have been associated with the question whether the text could
adequately address the ‘intelligible world’ at all (see Lucius and
Nicostratus ap. Simplicium in Cat. 73,15–76,16, with Boethus’
response, discussed further in ch. 4).
As alien as this kind of reading may appear now, it clearly did help
to inspire interest in the metaphysical and ontological meaning of the
Categories. Eudorus (like later Platonist readers of the Categories such
as Lucius) examines the Categories as concerned at least partly with
an account of beings, an ontology that can be ‘mapped’ to Pythagor-
ean and Platonic metaphysics. By contrast, Andronicus treated the
Categories primarily as a logical instrument, a competitor for the
throne of Chrysippean dialectic and not (as it seems) for Platonic
metaphysics.

3. The Categories may treat both intelligible

and perceptible objects
A third text about the Categories, Plutarch, On the Generation of the
Soul in the Timaeus 1023D–1024A, although not explicitly ascribed
to Eudorus, may strengthen the suggestion that Eudorus interpreted
the Categories in a Pythagorean framework. Regardless of whether
Eudorus is indeed Plutarch’s main source in that treatise, as has
sometimes been suggested, this passage occurs quite close to Eudorus’
discussion of Xenocrates and Crantor, and given Eudorus’ interest in
the scheme of ten categories, I think it is reasonable to consider its
candidacy as an imitation of Eudorus.25 This suggestion is quite
tentative, but I hope worth spelling out. (See Chiaradonna 2009b
for a very thoughtful and detailed exploration of Eudorus’ role in
this debate).
Plutarch is launching an inquiry into the soul’s capacity to appre-
hend sensible objects (yielding opinion and belief) by contrast to the
soul’s capacity for pure intellection (yielding real knowledge). In
order to illustrate that the soul of the cosmos is capable of judging
not only intelligible things, but also perceptibles (1023C), he cites
Plato, Timaeus 37A–B, which runs as follows:

25
On the passage and Eudorus’ role in Plutarch’s treatise, see for example Bonazzi
(2013: 163–70).
92 Eudorus of Alexandria
[Whenever the soul] comes into contact with something whose being is
scatterable or else with something whose being is indivisible, it is stirred
throughout its whole self. It then declares what exactly that thing is the
same as, or what it is different from, and in what respect and in what
manner, as well as when, it turns out that they are the same or different
and are characterized as such.26
Plutarch paraphrases this passage, and then adds the gloss that ‘in
these words, [Plato] is also giving an outline of the ten categories; he
makes himself even clearer in what follows’ (K  Ø – Æ ŒÆd H
ŒÆ ŒÆŪæØH Ø  ªæÆçc Ø Aºº E Kç B
ØÆÆçE, 1023E7–9). The passage mentions both divided and indi-
visible objects, and Plato himself goes on to state that this statement
about the soul’s declarations ‘applies both to the things that come to
be, and to those that are always changeless’ (37B2–3).27 The most
natural reading, I think, is that on Plutarch’s view, the declarations
that Plato attributes to the soul—about both partitioned (perceptible,
coming-to-be) objects and partless (intelligible, changeless) objects—
can be mapped to the scheme of ten categories.
Plutarch then paraphrases Tim. 37B–C, where Plato ‘makes him-
self clearer’. Here Plato has Timaeus state:
And when this contact gives rise to an account that is equally true
whether it is about what is different or about what is the same . . .
whenever the account concerns anything that is perceptible, the circle
of the Different goes straight and proclaims it throughout its whole soul.
This is how firm and true opinions and beliefs come about. Whenever,
on the other hand, the account concerns any object of reasoning, and
the circle of the Same runs well and reveals it, the necessary result is
understanding and knowledge. And if anyone should ever call that in
which these two arise, not soul but something else, what he says will be
anything but true.28

26
‹Æ P Æ ŒÆc å Ø KçÅÆØ ŒÆd ‹Æ I æØ, ºªØ
ŒØı Å Øa Å ÆıB ‹ø fi ’ ¼ Ø ÆPe fi q ŒÆd ‹ı i æ, æe ‹Ø 
ºØÆ ŒÆd ‹fiÅ ŒÆd ‹ø ŒÆd › ı Æ Ø ŒÆa a ªØª   æe ŒÆ
ŒÆÆ r ÆØ ŒÆd åØ ŒÆd æe a ŒÆa ÆPa åÆ I (ed. Burnet). The final
words (37B2–3) are omitted in Plutarch’s paraphrase.
27
While Plutarch does not paraphrase these words directly, I think that the
remainder of his discussion makes clear that he has the distinction between intelligible
and sensible being in mind.
28
ºª b › ŒÆa ÆPe IºÅŁc ªØª  æ  Łæ k ŒÆd æd e ÆP,
K fiH ŒØı ø
fi ç’ ÆF çæ  ¼ı çŁªªı ŒÆd MåB, ‹Æ b æd e ÆNŁÅe
ª ªÅÆØ ŒÆd › F ŁÆæı Œ Œº OæŁe Ng N AÆ ÆPF c łıåc Øƪª ºfiÅ,
 ‘Pythagorean’ Support 93
As Plutarch continues to explain this passage (1023F), the soul may
settle on the sensible (e ÆNŁÅ) in the ‘circle of the other’, estab-
lishing opinion and belief; but when the soul settles on reasoning (e
ºªØØŒe) in the ‘circle of the same’, knowledge is fulfilled
(KØ Å K IªŒÅ IºEÆØ). These two different motions of
the soul are then distinguished as apprehending ‘what is subject to
sense’ (IغÅØŒc F ÆNŁÅF) and an intelligible motion that
culminates in knowledge (æÆ B ÅØŒB KŒ Å ŒÆd ºıÅ
N KØ Å, 1024A).
The key question for us is what Plutarch means when he says
that Plato is offering an outline account or ªæÆç of the ten
categories in Tim. 37A–B, and that he clarifies his account further
in Tim. 37B–C. In particular, is the ten categories’ scope of applica-
tion meant to be limited to perceptible things, or does it extend also to
intelligible things? The most natural answer, I think, is that the ten
categories apply to the soul’s activity of predication in both ‘motions’.
Granted that the passage as a whole aims to show that the soul is not
restricted to apprehension of intelligibles, the point is not that the
soul is restricted to the apprehension of sensibles either; Plato is said
to be describing the soul’s two different motions, one underwriting
sensation and opinion, the other intellection and knowledge, and ‘in
these words’ (K  Ø, 1023E7) he is providing an account of the
ten categories.
If this passage does reflect not only Plutarch’s view but also
Eudorus’, it may suggest the following further interpretation of the
fragments of Eudorus in Simplicius. Eudorus may have considered
the ‘circle of the same’ as reflecting the kind of absolute predication
that could apply to the ten categories in and of themselves (for
example, predicating ‘animal’ of ‘human’ and ‘colour’ of ‘pale’),
while the ‘circle of the different’ reflected relative or accidental predi-
cation (for example, ‘pale’ of ‘Socrates’). Also, P Æ could always be
in some sense measured by Time and Place, and yet still be intelligible
P Æ, in intelligible Time and Place. This was also the view of
Iamblichus, who talks of ‘intelligible time’ and ‘intelligible space’,

 ÆØ ŒÆd  Ø ª ªÆØ ÆØØ ŒÆd IºÅŁE, ‹Æ b Æs æd e ºªØØŒefi q ŒÆd ›
F ÆPF Œ Œº hæå J ÆPa Å fiÅ, F KØ Å  K IªŒÅ
IºEÆØ·  ø b K fiz H Zø Kªª ªŁ, ¼  Ø ÆPe ¼ºº ºc
łıåc YfiÅ, A Aºº j IºÅŁb KæE (ed. Burnet).
94 Eudorus of Alexandria
and claims dependence on ‘Archytas’ for his interpretation; so it may
be reasonable to associate Eudorus with this view via Archytas. It is
certainly true that Archytas raises questions about whether the Cat-
egories deals with the sensible world, or which categories do apply to
the sensible and which to the intelligible (see 22,31; 30,19 Thesleff,
with Szlezák ad loc.), and that these are later taken up by Lucius and
Nicostratus.
Aside from this, the An. proc. passage could shed light on two
further points:
(a) Eudorus believed that Aristotle offered an inadequate account
of the non-substance categories, in that Aristotle fails to
describe them in their Absolute sense, only in their Relative
sense. This would not make much sense if Eudorus supposed
that the Categories was solely concerned with sensible being: it
seems unlikely that he would have argued that sensible Quality
was absolute or per se. But he might well have argued that
intelligible Quality has an ‘absolute’ sense that needs account-
ing for. Thus Eudorus might have treated the category of ‘per
se’ or absolute as corresponding to Aristotelian P Æ—includ-
ing intelligible being (cf. Bonazzi 2013: 182).
(b) Second, the later line of criticism against the Categories main-
tained by Lucius and Nicostratus suggests that the Categories is
not an adequate account of intelligible reality. This case would
be difficult to make for a critic who already assumed that the
Categories was not meant to be an account of intelligible reality
at all: if Eudorus influenced Lucius, then (as seems chrono-
logically probable), it appears unlikely that Eudorus restricted
the scope of the Categories to the sensible world, which would
make Lucius’ attack on it look quite misguided (see ch. 4). We
might alternatively suppose that Lucius followed Eudorus in
supposing that the schema of ten categories was itself a good
guide to both sensible and intelligible reality, but that Aris-
totle’s account was lacking. Later criticism certainly continued
to question the viability of the ten-categorial scheme as an
account of intelligible substance, if that is what it was meant to
do (through to Plotinus Enn. 6.1–3).
 ‘Pythagorean’ Support 95
Relationship to Andronicus
We have noticed a number of similarities between Eudorus and
Andronicus above, such as their common relabelling of Where and
When as Place and Time, their common reversion to the Academic
categories of Absolute and Relative as a useful umbrella for the ten
categories, and their common treatment of the æe  as an ‘offshoot’
or ‘appendage’ of the other nine categories. Did Andronicus respond
to Eudorus, or Eudorus to Andronicus, or neither?29 I have argued
above that Andronicus may well have introduced the treatise Cat-
egories to a wider readership, but Eudorus may have worked inde-
pendently on the doctrine of the ten categories, and both may
independently (or in dialogue and debate with one another) have
attempted to integrate and relate the two categories of absolute and
relative with the ten categories. Perhaps, again, since Xenocrates is a
common influence on Eudorus and Andronicus elsewhere in their
philosophical oeuvre, we might look to Xenocrates as a common
source for shared factors in their interpretations of the ten categories
(if not necessarily the text Categories), just as Boethus would turn to
Speusippus as a useful source in interpreting the ‘onymies’ (Simplic.
in Cat. 38,19–23), and Speusippus even appears, at one point, to have
essayed an interpretation of the opening of the Categories (29,5).30
Since the influence and novelty of Andronicus’ text are unclear (see
ch. 2 and Appendix 2), we do not have to assume that his work
preceded Eudorus’. And the early critic Lucius, and later Nicostratus
(see ch. 4), who seem to follow Eudorus in entertaining an ontological
or metaphysical reading of the Categories, show no obvious signs of
knowledge of Andronicus’ distinctive reading of the treatise:31 in fact,
if Andronicus’ introductory phrase (H ºª ø e b ¼ı
ı ºŒB ºªÆØ . . . )32 suggested verbal or logical subject-matter,

29
Dillon 1977/1996 also points out that Eudorus may have reacted to Ariston’s
commentary, in which case—if it is correct to give Andronicus tentative preference
over Ariston (see ch. 2)—Ariston might have mediated Andronicus’ interpretation to
Eudorus.
30
Fr. 68b Tarán = 32b Lang; see Chase 2003: 117 n. 310, Tarán 1978b.
31
As illustrated by the suggestion, referring to Simplic. in Cat. 1,19–20, that Lucius
is not criticizing a text that lacks the words B P Æ at Cat. 1a1–2, as Andronicus’
text (according to Simplicius) did.
32
Moraux 1973: 102–3 pieces together the full preamble of Andronicus’ para-
phrase from Simplicius in Cat. 21,22–4, 26,18–19, 30,3–5 and Dexipp. in Cat.
21,18–19. See also Sharples (2008: 280).
96 Eudorus of Alexandria
Lucius’ criticism of the Categories as failing to address intelligible
being at all seems to miss the mark.33 Perhaps most tellingly, Lucius
and Nicostratus claim that Aristotle opens by discussing æª ÆÆ
(Simplic. in Cat. 21,2–25,9); if they read the text through an Andro-
nican lens, under the influence of Andronicus’ paraphrase, this would
be a curious assertion to make (although given the fragmentary
nature of the evidence for Lucius and Nicostratus, little is certain).
Moreover, we know from Simplicius that Andronicus’ text (or para-
phrasis) of the Categories did not include the full phrase ºª B
P Æ at Cat. 1a1–2, but included only the word ºª. But Nicos-
tratus, likely ‘taking over’ a point of Lucius (ƺº , cf.
Simplicius in Cat. 1,19–20), criticized Cat. 1a1–2 as suggesting that
homonymy only applies to the first category (29,23–30,5). Consider-
ing these points, it seems at least plausible that Lucius and Nicostratus
were not reliant on Andronicus’ presentation of the Categories,
but worked from (something more like) Eudorus’ interpretation of
the text.
In fact, we might treat Andronicus’ choice to reframe the text by
the introductory phrase H ºª ø . . . , together with his choice
to introduce the curriculum with the Categories as a work of logic, as a
reaction to the kinds of problems raised by Eudorus and (perhaps
later) Lucius—namely, problems about the function of the Categories
as a work about beings. By asserting that the Categories was a work
about words, and publishing it with a ‘paraphrasis’ that supported this
assertion, Andronicus might have addressed those problems, as
Boethus later did (ch. 6), while paving the way for new (or resurrected)
issues about its inadequacy as a work about language, raised by Stoics
like Athenodorus and Cornutus (cf. ch. 5). These Stoic problems may
in turn have necessitated the response of Boethus. On such a scenario,
one could see Andronicus as reactive to the Platonizing or Pythag-
orean reading of the Categories in Alexandria—taking over from
Eudorus elements like the overarching focus on the Absolute and the
Relative, among other factors.34 At the same time, he clearly resituated
the Categories as a work of ‘logic’ that introduced the Organon, even as

33
Lucius and Nicostratus clearly grant (Simplicius 73,15–76,16) that the text is
about beings when they inquire whether it concerns both intelligible and sensible
beings.
34
Such as the replacement of ‘When’ and ‘Where’ with ‘Time’ and ‘Space’,
discussed above.
 ‘Pythagorean’ Support 97
important features of Eudorus’ reading—such as the reliance on the
Sophist’s distinction of absolute from relative as the ‘overarching’
division under which the ten ŒÆŪæ ÆØ fell—were carried across
into his interpretation.
When Eudorus evinced interest in the Categories, he appears to
have represented Aristotle as a genuine, if sometimes errant, exposi-
tor of a Pythagorean world view, a picture of how the world was
actually structured. We may add that some later Neoplatonic sources,
such as Elias [David] in Cat., refer to Aristotle as a ‘true Pythagorean’
(Puthagoreiōi kat’ alētheian gegonoti, 107,6 Busse), a claim which is
echoed in the anonymous Life of Pythagoras, where Aristotle is
described as the tenth diadokhos of the Pythagorean school, following
Archytas and Plato. Such reports could reflect a desire on the part of
earlier, first-century Pythagoreans to ‘appropriate’ Aristotle into the
Pythagorean tradition, as some (such as Bonazzi 2007) have sug-
gested; if so, this movement could have played an important role in
the birth of interest in the Categories, a text which Eudorus and the
authors of the Archytan treatises clearly saw as susceptible of Pythag-
orean exegesis.

II. PSEUDO-ARCHYTAS (FL. c. 50 bce?)

Archytas of Tarentum was a Pythagorean contemporary of Plato,

who appears to have studied a number of specialized problems in
music, mathematics, and mechanics. His association with the Acad-
emy is documented in the Seventh Letter ascribed to Plato;35 accord-
ing to Philodemus’ Index Academicorum (col. 6, 12) he was Plato’s
student. While the works of Archytas himself were subsequently lost,
centuries later a wide array of Pythagorizing texts were published
under the name Archytas, evidently in a bid to associate a number of
key doctrines with the ‘ancient’ Pythagorean school.36 The bid was
largely successful, certainly by late antiquity; with the exception of
a notice in Themistius (ap. Boethium in Cat. PL 64, 162A), the

35
On which see Lloyd 1990.
36
For what follows, see the introduction to T. A. Szlesák (1972). I have also found
the synopsis in Dillon (1977) and Bonazzi (2007) very useful.
98 Pseudo-Archytas
Pseudo-Archytan works seem to have found wide acceptance as
genuinely ‘Pythagorean’.37
Several of these pseudepigrapha focused on the Categories: (I) ‘On
the Universal Account’ (peri tou katholou logou); (II) ‘The Ten
Universal Statements’ (katholikoi logoi deka, which is evidently a
much later, and simpler, text); (III) ‘On Opposites’ (peri antikeime-
nōn). The first two texts are extant (ed. Thesleff; see Szlezák, 1972);
the third is lost, but preserved in fragments in Simplicius. All that we
now possess in full is a koinē version of the first and substantial
fragments in Doric of all, the latter preserved by Simplicius and
other later commentators.38
Here I would like to review very briefly the place of Pseudo-
Archytas in the early discussions that we have been surveying so
far, simply with a view to showing that he also represents a broadly
favourable, ‘Pythagoreanizing’ interpretation of Aristotle’s Categories
(and categories) which does not aim at rejecting them, but at exploit-
ing them as a valuable resource on the road to knowledge (cf. Bonazzi
2013: 179–83). Szlezák (1972) argued that Pseudo-Archytas is react-
ing to the developments of the first generation of commentators,
beginning with Andronicus. For example, ‘Archytas’ assumes that
the ‘categories’ concern only the sensible world (22,31 and 31,5
Thesleff: cf. Simplic. 76,20), which may differ from Eudorus’ inter-
pretation. He separates the last portion of the Categories, the post-
praedicamenta, from the rest of the text in (I) above, but (III)
examines them as Pythagorean; as Frede points out (Der Neue
Pauly), it seems unlikely that a single author is behind both (I) and
(III). ‘Archytas’ also agrees with Andronicus and Ariston in improv-
ing on Aristotle’s definition of the Relative (Simplic. 202,2).
Is Pseudo-Archytas also drawing on the work of Eudorus?39 As we
have seen in our brief review of Eudorus, this is almost certainly the
case; for example, ‘Archytas’ provides similar Pythagorean reasoning

37
Indeed, by the fourth or fifth century bce, ‘Pythagorean’ effectively meant
something that modern scholars tend to call ‘Neoplatonic’; see for example the
proem to Syrianus in Met.
38
The fragments of the latter are edited by Thesleff 1965: 15–18; the first by
Thesleff 1965: 21–32 and Szlezák 1972: 34–57. The latter prints the two versions
separately. The katholikoi logoi deka is edited by Thesleff 3–8 and Szlezák 61–8.
39
So Dillon (1977: 134 and 2000) suggests, alongside Szlezák and Theiler. Con-
versely, Dörrie 1976: 300 suggests that Eudorus is opposed to Aristotelian innovations
in Platonism.
 ‘Pythagorean’ Support 99
for the order of the categories (cf. Szlezák 1972: 116). But there are
differences: ‘Archytas’ puts Relation fourth in the order of categories,
after Quality and Quantity, and Space and Time at the end of the list.
Philo of Alexandria, perhaps drawing on Eudorus, does the same.
Several points worth stressing here are ‘Archytas’’ view that the
Categories concerns sensible entities (cf. 31,3–5 Thesleff: ‘Æ ªaæ
ÆFÆ çıØŒA P Æ ŒÆd ø ÆØŒA ı  Æ KØ, Iºº’ P ÆA
ŒÆd IŒØø ŒÆd æØ ª I æ’), and his view that man is the
measure of knowledge ‘because he is endowed by nature with the
capacity to enumerate the principles by means of which all things
are known’ (31,32–32,33). All knowledge, as ‘Archytas’ goes on to
explain, seeks to apprehend the infiniteness of reality by means of
principles limited in number; as all number is in the number ten, so
there should be ten kinds of being and predication, and human beings
have ten fingers on which to count them. These arguments, or ones
very much like them, are later ascribed to Herminus: both reality and
words are infinite, and we use the ten predications to help apply limit
to the unlimited (Porphyry, in Cat. 58,9–20). Our capacity to apply
such a limit comes through our power to enumerate (IÆæØŁ E) the
ten modes of predication (Simplicius, in Cat. 62,7–23). I have sug-
gested above that Andronicus may have built on the same central
ideas; and it is clear from Pseudo-Archytas that these notions go back
well before Porphyry.
Thus both Eudorus and Pseudo-Archytas are prepared to integrate
the ten categories into their own philosophy as an ontological theory,
not merely as a dialectical or logical one. They also seem, based on our
limited evidence, to have made some direct use of our text of the
Categories to develop that speculation. This attitude lies in contrast to
Lucius and Nicostratus, who also spring from a Platonist background,
but focus on developing criticisms of the Categories, ranging from
doubts about its scope to criticisms of its organization and content.
But I hope to have shown that Eudorus and Pseudo-Archytas devel-
oped a plainly metaphysical reading of the text, treating it as a useful
account of being, and that this in turn may have grounded the kind
of interpretation that Lucius and Nicostratus (ch. 4) felt it necessary
to rebut.
 Part B

Early Criticisms: Platonists and Stoics

 4

Lucius and Nicostratus

In this chapter, I argue that Lucius, a critic of the Categories named

by Simplicius, criticized the Categories as an inadequate treatment
of ontology, and that his criticisms attracted a response from
Boethus of Sidon (see ch. 6). I also attempt to show that the
criticisms offered by Lucius and Nicostratus—two figures often
conflated by Simplicius—can be at least provisionally distinguished
from one another.
Others were content to write only puzzles (IæÆØ) against what
is said: this is what Lucius did, and after him Nicostratus, who
appropriated the considerations of Lucius. These two vied with
each other in providing objections (K Ø ) to nearly everything
said in the book, and they did not go about their task with respect,
but rather in a violent and shameless manner. Nevertheless, we must
be grateful to them, too, both because the puzzles they set forward
had, for the most part, a basis in fact (æƪ Æ Ø Ø ), and because
they provided their successors with starting-points both for the
resolution of the puzzles, and for the development of many other
excellent theories.1
Simplicius, in Cat. 1,18–2,2 (tr. Chase 2003, lightly adapted)

1
¼ººØ b Xæ   IæÆ Æ ªæłÆØ æe a º ª Æ, ‹ æ ¸ŒØ 
 ÅŒ ŒÆd ’ ÆPe ˝ØŒæÆ a F ¸ıŒı ƺº  , å  Ø
æe Æ a NæÅ Æ ŒÆa e غ K Ø Œ Ç Ø çغØ  Ø, ŒÆd P b
PºÆH , Iººa ŒÆÆçæØŒH Aºº ŒÆd IÅæıŁæØÆŒø · ºc ŒÆd Ø åæØ , ŒÆd
‹Ø æƪ Æ Ø Ø a ººa H IæØH æ º ŒÆd ‹Ø º  H
IæØH Içæ a ŒÆd ¼ººø ººH ŒÆd ŒÆºH Ł øæÅ ø E Ł’ Æıf
K ŒÆØ (ed. Kalbfleisch 1907).
104 Lucius and Nicostratus
Lucius2 and Nicostratus3 are shadowy figures, known to us through
their recurring role as antagonists in the pages of Simplicius’ com-
mentary on the Categories. They are made to portray the antithesis of
Simplicius’ sketch of the capable exegete of Aristotle (7,23–8), who
exercises impartial criticism (ŒæØ I ŒÆ) by homing in on the
essential agreement of philosophical schools (ı çøÆ ÆPH
IØå  Ø) without indulging in ‘disputatious twaddle’ (cf. KæØØŒc
çºıÆæÆ, 8,1). By contrast, Nicostratus and his followers are repre-
sented as eager polemicists (çغ ØŒFØ, 268,19) who deploy
Platonic doctrines and Stoic linguistics as sticks with which to beat
Aristotle, and exploit the letter of the text without regard for its
deeper intent (F : cf. 7,31).4 They accuse Aristotle of ‘useless talk’
( ÆÆغªÆ, 58,15), charge him with ignorance (ÆNØAÆØ . . .
Iª E, 231,21), and ‘waste time’ (370,1) with their ‘accusations’
(406,6; 410,25), managing to ‘bury the truth’ completely (370,8).
Their errors are ‘easy’ to recognize (Þfi  Ø ıæA, 428,3).
There are no direct references to ‘Lucius’ and ‘Nicostratus’ outside
Simplicius in Cat.5 Even in those cases where their criticism of
the Categories appears to be implicit in other texts—for example,
in Plotinus, Enneads 6.1, Dexippus, and Porphyry’s commentary
by Question and Answer—we are reliant on Simplicius’ testimony to

2
We know of Lucius only from eight references in the pages of Simplicius in Cat.
On him see Capelle 1927, Praechter 1922 = 1973: 101–37, and more recently the
excellent entry in Goulet (2005) by Concetta Luna, as well as Dillon 1977: 344–5, and
Strange 1987: 956 n. 8. The difficulty of distinguishing Lucius from his apparent
successor, Nicostratus, has been noted by Gottschalk 1987: 1104; see also Moraux II:
528–63. His floruit is typically dated to the second century ce, but Sedley 1997: 117
n. 26 points out that the evidence is very loose.
3
Simplicius describes Nicostratus as ‘following’ Lucius and as ‘taking over’ his
positions (1,19). On him see recently Gioè 2002: 155–219; Dillon 1977: 233–6;
Praechter 1922 = 1973: 101–37; K. von Fritz, ‘Nikostratos 26’, RE 17.1 (1936), cols
547–51; Moraux II: 528–63. Nicostratus lived in Athens, and Praechter identifies him
persuasively with the Nicostratus honoured in an inscription at Delphi (Sylloge II3
Nr. 868). See also Zeller 1990: III.1 [4], for whom Nicostratus was a Stoic.
4
In addition to the commentator’s force of the ‘sense’ or ‘point’ of the language
(LSJ s.v.  ), this remark may echo the ‘noeric’ interpretation, or higher criticism,
that Iamblichus is credited with introducing to the study of the Categories. On
Iamblichus’ noeric analysis, see for example Dillon 1997: 65–77, I. Hadot’s commen-
tary on Simplicius 2,11 and related samples of Iamblichus’ criticism in Commentaire
sur les Catégories I–III (Leiden, 1990), and the introduction to Michael Chase’s
translation of Simplicius in Cat. 1–4 (2003). Sorabji (2005c), Vol. 3, very usefully
summarizes the relevant passages.
5
As Moraux 1984: 530 points out, this compels us to acknowledge the exclusive
influence of Simplicius in selectively shaping their representation.
 Lucius and Nicostratus 105
identify them. As we have them, then, ‘Lucius’ and ‘Nicostratus’ are
essentially constructs of Simplicius’ text.
It is widely held that the views of Lucius and Nicostratus cannot be
usefully distinguished from each other within the text of Simplicius.6
This consensus is certainly justified by the limited nature of the
primary evidence. In the course of this chapter, however, I shall
argue that passages ascribed to ƒ  æd e ¸ŒØ consistently betray
a different set of concerns from those ascribed to Nicostratus, and
that these two sources, as Simplicius represents them, may be usefully
separated and discussed in isolation. I shall maintain this position
through a close analysis of the identifying cues that Simplicius uses to
introduce and contextualize their views. For this purpose, a key part
of this chapter (which will also be useful in following chapters) is the
Table of Responses, showing the (historical or constructed) ‘dialogue’
between aporists and defenders of the Categories. (For a more detailed
treatment, see Griffin 2009: ch. 3).
The approach of this chapter is therefore to explore the manner in
which Simplicius portrays these two figures and their polemical
relationships with other sources, including Boethus, Plotinus, Por-
phyry, and Iamblichus.7 Primarily, the present chapter aims to argue
that the views of Lucius—which challenge the value of the Categories
as a treatise in ontology or first philosophy—are answered by Boethus
of Sidon. I construct a typology of these references in the main body
of the paper, and attempt to establish several conclusions about the
differing interests of the ‘Lucius’ source and the ‘Nicostratus’ source,
and to characterize the anonymous and named sources who respond
to them. Specifically, I argue that ƒ  æd e ¸ŒØ are primarily
concerned about ‘doctrinal’ points in the text, involving (for example)

6
Moraux 1984: 528–63 stresses the difficulty of usefully distinguishing these two
sources. Hoffmann and Hadot (1990: 4 n. 11) point out that ‘il lui paraît impossible de
distinguer la pensée de Lucius de celle de Nicostrate’. Gottschalk (1987: 1104) notes
that Lucius as a persona ‘is only known through his association’ with Nicostratus, a
point that I acknowledge and expand in my discussion of historicity below.
7
This chapter is thus primarily concerned with the identification and distinction
of sources as represented by Simplicius. While it is not my primary concern to
evaluate the validity of Lucius and Nicostratus’ views as such, or their usefulness as
an interpretation of the relevant passages in the Categories, I have endeavoured to
provide cross references to more detailed discussions of certain passages in other
chapters.
106 Using the Neoplatonist Sources
the distinction of intelligible from sensible reality. The Nicostratans,
on the other hand, demonstrate a broader set of textual, rhetorical,
and logical concerns, which seem to have interested Atticus
(30,16–22) as well as Porphyry and Iamblichus.

I. USING THE NEOPLATONIST SOURCES

In the course of this study, I shall analyse the identifying cues that
Simplicius uses to introduce and contextualize the views of Lucius
and Nicostratus. There are just four: (1) ƒ  æd e ¸ŒØ; (2) ƒ
 æd e ˝ØŒæÆ; (3) › ˝ØŒæÆ ; (4) ƒ  æd e ¸ŒØ b
ŒÆd e ˝ØŒæÆ.
The name of Lucius appears in eight contexts throughout the first
third of Simplicius in Cat. (1,19; 48,1; 62,28; 64,18; 73,28; 125,16;
127,30; 156,17), and abruptly ceases to be cited after Simplicius’
discussion of Quantity. There is no singular › ¸ŒØ source (except
for the passage in the proem cited above), and even verbatim quota-
tions are often ascribed to a plural collective of ƒ  æd Lucius or ƒ
 æd Nicostratus. (The locution ƒ  æd e X might, of course, simply
refer to X as the originator of a tradition rather than a literal
‘school’8—although, as I suggest below, its later ancient use need
not necessarily imply the existence of an historical figure). Nicostra-
tus is cited more frequently, on thirty occasions balanced throughout
the entirety of the in Cat. Like Lucius, he is often the focus of a school
of followers (ƒ  æd e ˝ØŒæÆ), but he is also a lone speaker
on many occasions (e.g. ˝ØŒæÆ KªŒÆº E, 370,1; X Ææ 
˝ØŒæÆ , 381,23; Iغª Ø ˝ØŒæÆ , 390,15). He is also
cited alone ([› ˝ØŒ]æÆ ) in the new fragmentary commentary,
plausibly attributed to Porphyry, in the Archimedes Palimpsest
(Chiaradonna, Rashed, Sedley, and Tchernetska 2013, 2,26–31).
Regardless of the introductory cue, Simplicius frequently continues
a ‘dialogue’ argument simply with ‘them’ (çØ, with the speakers’
identity carried over from previous citations), even if the first cue is

8
On the locution itself, see Radt 1988: 35–40. I am indebted to Prof. Dirk Obbink
for drawing this reference to my attention.
 Lucius and Nicostratus 107
singular; he seems to have little interest in drawing a distinction
between the parties. He is even content to roll Lucius, Nicostratus,
and Plotinus together as a single interlocutor, as at 73,28 and
following.
Simplicius traces certain of Plotinus’ arguments in Enneads 6.1
directly to Lucius and Nicostratus. In many cases, the stratigraphy of
Simplicius’ text clarifies the places where Plotinus picked up aspects
of their argument and developed their thought, and this sheds add-
itional light on their interests.9 Simplicius has Plotinus endorse,
and sometimes spearhead, the criticisms ascribed to Lucius and Nicos-
tratus (for example at 127,11–128,10). He describes their criticisms in
similar language: for example, Plotinus provides a æƪ Æ Øø -
Æ K  Ø (2,3), whereas Lucius and Nicostratus deliver
æƪ Æ Ø Ø considerations (1,22) alongside their polemics. Simpli-
cius sometimes appears to contrast such æƪ Æ Ø Ø exegesis with
the ‘noetic’ approach ascribed to Iamblichus (at in Cat. 2,9 and follow-
ing; see also Dillon, 1997). I note each of the explicit correspondences
with Plotinus, and some implicit correspondences, below.
Lucius’ and Nicostratus’ ‘schools’ are explicitly cited by name
throughout the first third of Simplicius’ commentary, appearing as
a source for the discussions of Substance, Quantity, and Quality. But
the name of Lucius ceases to be cited after the discussion of Quantity,
and his last appearance is on page 156 of Kalbfleisch’s text. Nicos-
tratus and his ‘school’ continue to be cited by name throughout
Simplicius’ entire commentary, and are distributed throughout his
discussions of the ten categories.

9
It remains an open problem whether Plotinus repeats the assertions of Lucius and
Nicostratus in Enneads 6.1 in order to criticize the scheme of ontological genera that
he takes the Categories to present, or simply in order to show that the Categories does
not purport to describe a scheme of ontological genera at all (contrary to the view of
some commentators), and may therefore be compatible with the scheme of genera
that Plotinus himself will present for the intelligible world and for the sensible world
in 6.2–3 respectively. Both positions were recently sketched by de Haas and Chiar-
adonna in Celluprica and D’Ancona (2004, with earlier bibliography in both articles);
Chiaradonna maintains the ‘differenza netta tra le tesi di Plotino e Porfirio’ against the
arguments of Strange (1987) and de Haas (2001), whereas de Haas maintains that ‘the
echoes in Plotinus of . . . Lucius and Nicostratus do not testify to his agreement with
the critical tradition, but to his exploitation of it for his own purposes.’ See also
Chiaradonna (2002) on the earlier origins, perhaps with Lucius and Nicostratus, of
Plotinus’ arguments against the Aristotelian genus.
108 Life and History
II. LIFE AND HISTORY

Lucius
Excluding Simplicius’ notices in his Categories commentary, there is
no evidence whatsoever for Lucius as an historical figure. In fact, even
Simplicius’ direct references cite a school of associates (ƒ  æd e
¸ŒØ) rather than an individual.10 Moraux notes that no known
Imperial philosopher can be identified with Simplicius’ ‘Lucius’.11
Dillon (1977/1996: 344–5) observed the ‘remote possibility’ that
Simplicius’ Lucius should be identified with the Etruscan Pythagor-
ean represented by Plutarch (QC 8.7–8) as active during the nineties
ce. However, if we do accept that Lucius was a real person, I suspect
that he flourished before the death of Boethus, who is represented by
Simplicius as ‘replying’ to the embarrassments presented by Lucius’
school; given the evidence that makes Boethus a contemporary of
Strabo, born around the sixties bce or earlier (Strabo, Geog. 16.2.24),
it would be a difficult stretch to extend the activity of his historical
interlocutor into the nineties ce. It is also interesting that Simplicius
does not mention Lucius among the ‘ancient exegetes’ of the Categor-
ies (in Cat. 159,32–3), although there are various reasons that Lucius
might not appear in that list; he may not have been viewed by
Simplicius or his source as an KŪÅ of the text as such, or he
may simply not have adopted the view on the relative that Simplicius
is ascribing to the other ‘ancient exegetes’, namely Boethus, Ariston,
Andronicus, Eudorus, and Athenodorus.
On the other hand, I have been tempted to speculate that this
Lucius was not originally a single historical figure carrying this
name; perhaps the name served, in some philosophical or doxogra-
phical work, to represent a broadly Pythagorean or Platonist school of
thought, or was a pseudonym representing another person whose
name we know better. For example, if Boethus of Sidon composed a

10
Of course, the phrase ƒ  æd e ¸ŒØ may simply convey the force ‘Lucius
and his followers’. It is noteworthy, however, that Simplicius varies the personal name
with the ƒ  æd X locution often in the case of Nicostratus (contrast Simplicius in Cat.
30,16–22; 231,19–23; 368,1–371,27; 385,9–15; 388,1–13; 390,14–19; 402,11–403,5;
406,5–407,14; 410,25–411,28; 414,23–415,15; 428,3–13; 429,13–430,4 against [with
ƒ  æd] 21,2–25,9; 26,21–28,1; 29,23–30,5; 257,31–6; 268,19–32), but never in the case
of Lucius.
11
Moraux 1984: 529.
 Lucius and Nicostratus 109
commentary in ‘question and answer’ format, which could have been
an indirect model for Porphyry’s own commentary ŒÆa  FØ ŒÆd
IŒæØØ, it might have featured an inquisitor of this name. The
name Lucius itself might imply literary origins, and would be a
suitable fictive name to convey Pythagorean or Platonic associ-
ations.12 Such an interlocutor’s ‘questions’ (which may have extended
only through the discussions of Substance, Quantity, and Quality, for
his name is no longer cited by Simplicius after in Cat. 156,17) might
then have been taken up and developed by—among others—the
historical philosopher Nicostratus in the second century.13 But
there are problems with this hypothesis.14

12
‘Lucius’ was a common name for protagonists in contemporary literature, and
might even be viewed as a Roman ‘Smith’ or ‘Jones’: see for example H. J. Mason, ‘The
Distinction of Lucius in Apuleius’ “Metaphorphoses’ ” ’, Phronesis 37.2 (1983),
135–43. For a sophisticated audience, the name might have carried some particularly
Pythagorean (and Platonic) connotations. The passage of Plutarch (Table Talk 728B)
mentioned earlier is one example: Plutarch’s ‘Lucius’ lives in accord with strict
Pythagorean symbola, and follows Moderatus of Gades.
13
Although purely speculative, such an hypothesis could be carried further: for
example, the name ‘Lucius’ might have masked a better-known contemporary or
predecessor of Boethus, or collected an agglomeration of real and invented tenden-
tious positions that Boethus set out to confute, much as Porphyry’s interlocutor
functions in his commentary By Question and Answer. If ‘Lucius’ indeed referred to
a fictional figure, or masked a better-known contemporary of Boethus, his literary
origins could easily have been lost to Simplicius, who seems to have had only indirect
access to the text of Boethus, and may have simply followed the habit of his sources—
such as Porphyry’s lost commentary Ad Gedalium and Iamblichus’ lost in Cat.—in
referring to Lucius by name. I also discuss this possibility in the chapter on Boethus.
14
For one, as Marwan Rashed has pointed out to me, it would seem very odd for a
Roman name to be chosen to represent a philosopher of any persuasion in the first
century ce. A person could, of course, happen to have a Latin name and do philoso-
phy: for instance, Plutarch several times mentions Publius Nigidius Figulus as a
philosopher (Cicero 20.3, An seni resp. gerend. Sit 797D). But the name ‘Lucius’
would hardly have been a natural indicator to take from the air without an historical
referent. It may be worth noting that Plutarch identifies a ‘Lucius’ as an Etruscan
student of the Pythagorean Moderatus of Gades in Table Talk 7 (727B; cf. Dillon
1997/1996: 345), and to some extent goes on to caricature this Lucius’ strict inter-
pretation of Pythagoreanism. Plutarch may have chosen this name and historical
identity to represent literalist Pythagoreanism because he had met Lucius and Lucius
had made an impression upon him, or because Lucius’ ideas were reasonably well
known (although he is not otherwise discussed, as far as I know); and as Dillon (1977/
1996: 345) points out, there is no reason other than the name to associate this Lucius
with the commentator on the Categories. The passage does at least provide a link
between the name and Pythagoreanism, and Moderatus, this Lucius’ teacher, may
have taken some interest in the doctrine of ten categories following Eudorus (see ch. 3
below). But this is all speculation.
110 Life and History
Regardless of Lucius’ historicity or identity, a locution like ‘the
followers of Lucius’ (ƒ  æd e ¸ŒØ) may refer most generally
to the group of doctrines that were growing in prominence with the
Neo-Pythagorean and Platonic revival of the early first century
ce,15 and to which Plotinus (who took up these arguments in
Enneads 6.1–3)16 was clearly sympathetic. This ‘school of Lucius’
was active in the lifetime of Boethus, as Simplicius explicitly
describes Boethus’ responses to their criticisms.17
These doctrines may be recognized in the critical concerns of
‘the Lucians’ in the pages of Simplicius; they are not so prominent
among ‘the Nicostratans’, as I seek to show below. The Lucians are
concerned with the possible failure of the ten categories to distinguish
the intelligible and sensible worlds (73,15–76,6), with Intelligible
Number (129,27), and with the substantiality of the Monad and
the Point (65,13–14), which they suspect Aristotle fails to recognize.
They wish to show that Quality is proximate to ousia, and is more
descriptively useful than the Relative in the sense that Socrates is
better described by his qualities than his relationships (156,16, cf.
49,31–50,3).18
Unlike Nicostratus and his ‘school’, it is difficult to contextualize
the ‘Lucians’ more precisely than this.

15
On the doctrinal context in Rome, see now Hadot 2007: 119–200.
16
In the following sections, I note Plotinus’ views as they are reported by Simpli-
cius, especially where these reports are relevant to Simplicius’ account of the aporiai
set forth by Lucius and Nicostratus. Although a full discussion of Enn. 6.1–3 falls
outside the scope of this chapter, I have summarized in my opening chapter (1) my
position on Plotinus’ status as a source, and specifically the problem of his supposed
hostility to the Categories and reliance on Lucius and Nicostratus (see de Haas and
Chiaradonna in Celluprica and D’Ancona 2004).
17
In general, Simplicius seems to position the ‘dialogues’ he portrays in chrono-
logical order, but he sometimes interjects strikingly anachronistic discussions, such as
the first-century Boethus ‘replying’ to problems raised by the third-century Plotinus.
I have sought to show in the preceding chapter that those passages concerning
Boethus, Lucius, and Nicostratus do represent a genuine historical discussion.
18
See Moraux (1984: 548–50), and further discussion below.
 Lucius and Nicostratus 111
Nicostratus
In contrast to Lucius, Karl Praechter has presented attractive evidence
for the historical Nicostratus,19 identifying him with the Claudius
Nicostratus who is honoured on a Delphic stone (Sylloge II3 868)
shortly after ‘Gaius, son of Xenon’ and shortly before the philosopher
Taurus. As this Gaius is probably to be identified with the Platonist
Gaius, and Taurus was another prominent philosopher of the Pla-
tonic school, we may conjecture with Dillon that Nicostratus was ‘in
more or less the same position as Taurus was later’, and we may also
follow Dillon in ‘seeing in all this a tendency on the part of the
Delphians (perhaps as a result of the influence of Plutarch) to honour
prominent Platonists’ (1977: 234). There was not likely to have been
any formal ‘Academy’ of which Nicostratus was ‘head’, but there
would be reasons, judging solely from his associations on this stone
and his attitudes toward Aristotle and the ‘intelligible world’ in the
pages of Simplicius, to identify him as a ‘Middle Platonist’. Con-
versely, some aspects of his methodology point to a Stoic background,
although he seems to use Stoic language merely as a polemical device
to trip up Aristotle (see for example his discussion of homonymy at
26,22): it is certainly not clear whether he was committed to any Stoic
doctrines.
Simplicius makes it plain that Nicostratus postdates Boethus (cf.
23,29 and following), whose floruit may have extended into the early
first century ce, and that he antedates Atticus (cf. 30,16–17), whose
floruit fell in the 170s ce. Thus the textual evidence supports a date
for Nicostratus as flourishing close to the early second century ce, as
the Delphic inscription honouring Claudius Nicostratus would
indicate.
Nicostratus’ doctrinal concerns appear less prominent in Simpli-
cius’ pages than are those of the ‘school of Lucius’. Nicostratus
is interested in raising puzzles (aporiai) about the structure of
Aristotle’s text (see 21,2; 24,6; 25,9), his style and rhetoric
(58,14–28), and imputed textual inconsistency (368,1–371,27). By
contrast, the ‘Lucians’ often criticize Aristotle’s positions because
they do not see these positions as complying with certain a priori
doctrines—for example, the setting of the intelligible world and the
substantiality of intelligible Number and Quality (73,15–76,16). But

19
Praechter 1922 = 1973: 101–37.
112 Distinct Portrayal as Critics
Nicostratus and his school seek to exploit perceived weaknesses in the
internal consistency of Aristotle’s logic or the text (e.g. 30,16–22); an
excellent example of Nicostratus’ own method is the long, evidently
verbatim, passage from Nicostratus (supported by Plotinus, and
rebutted by Simplicius and Iamblichus) at 368,1–371,27.

III. DISTINCT PORTRAYAL AS CRITICS

Simplicius paints Lucius and Nicostratus primarily as polemical

figures, creators of ‘puzzles’ or of ‘questions’ (IæÆØ). Simplicius
initially describes them as ‘content’ (Xæ  ) to be aporists: ‘some
were content to write only aporiai against what is said: this is what
Lucius did, and after him Nicostratus’ (1,19).20 In summary, Simpli-
cius commonly describes the activity of the ‘Lucians’, the ‘Nicostra-
tans’, and Nicostratus himself through the aporetic verb IæÇø or
through the polemical verbs KªŒÆºø and ÆNØ ÆØ.
Nicostratus and his ‘school’ are represented, especially early in the
commentary, as interested in a common set of rhetorical and logical
issues. In the examples that we have cited, the ‘Nicostratans’ suggest
that the text is poorly structured (21,2–25,9), that Aristotle’s defin-
itions of homonymy, synonymy, and paronymy in Categories ch. 1
are insufficiently precise or self-contradictory (26,21–8,1, 29,23–30,5,
30,16–22), that Aristotle is being too wordy (58,14–28), and that he is
being sloppy and inconsistent in describing his own key terms,
such as ‘having’ (368,12–369,14). They find fault with Aristotle’s use
of illustrative examples for various reasons, sometimes on grounds
of style (58,14–28), sometimes on grounds of appropriateness
(429,13–430,4). Elsewhere, in the newly discovered fragment from a
commentary that may be attributed to Porphyry,21 Nicostratus raises
a familiar puzzle about the ‘transitivity principle’ at Cat. 1b10–11:22 if
Socrates is a man, and man is a species, then the transitivity principle
yields the absurd conclusion that Socrates is a species. Here as

20
I have explored their representation in more detail in Griffin 2009: ch. 4.
21
Chiaradonna, Rashed, Sedley, and Tchernetska 2013, 2,26–31.
22
If A is predicated of B as of a subject, then any C predicated of A will also belong
to B. For example, if man is predicated of Socrates as of a subject, and animal is
predicated of man, then animal is also predicated of Socrates.
 Lucius and Nicostratus 113
elsewhere, Nicostratus is pointing out inconsistencies and problems
in the formulations used by the Categories’ author. In general, Nicos-
tratus and the ‘Nicostratans’ demonstrate a broad interest in the
Categories as a treatise, tackling the minutiae of many categories
such as Having (as Plotinus appears to take some cues from Nicos-
tratus in 368,1–371,27) as well as the theory of contraries and contra-
dictories. Their criticisms occur throughout the text, albeit with a
definite bias towards the first three categories, which may be a bias of
Simplicius’ intermediary sources rather than their own bias.
The ‘Lucians’, on the other hand, appear to approach the Categor-
ies with a more specific set of doctrinal concerns in mind. They are
disturbed about the demotion of Forms and Qualities to the status of
accidents, and strive to reduce this position to absurdity (48,1–50,3);
in general, perhaps following the view adopted by Eudorus, they find
that qualities are undervalued in Aristotle’s order (156,16). They
argue that the system of the Categories is insufficiently rich to encom-
pass intelligible reality as well as sensible reality (73,15–76,16). They
find no place in Aristotle’s system for the Monad and the Point, those
essential Pythagorean entities, and roundly criticize Aristotle for
this omission (65,13–14). (Notably Boethus, in his response to the
Lucians, is represented by Simplicius as accepting a certain substan-
tiality for the Monad:  a Ø,  b PÆ,  K fiH ÅfiH
IæØŁ fiH . . . , 65,20–1).23 The Lucians are also accustomed to a philo-
sophical framework that establishes ‘body’ (H Æ) as substantial, and
they are puzzled to find it demoted to the status of a Quantity
(125,16). All of this suggests that the Lucians ascribed some onto-
logical value to the ordering of the ten categories, and commenced a
tradition of criticism that would continue in Plotinus 6.1.

IV. QUESTIONS AND RESPONSES

The puzzles put forward by Lucius and the Nicostratans are never
final. Simplicius presents their criticisms and questions, and then
describes the various solutions provided by his predecessors—

23
This well-known and somewhat controversial passage is discussed in detail in
my chapter on Boethus below, specifically with reference to the debate of Moraux and
Tarán regarding Boethus’ position on intelligible number.
114 Questions and Responses
especially Boethus, Porphyry, and Iamblichus. As a result, we are left
with the strong impression that the schools represented by ‘Lucius
and Nicostratus’ sparked a long tradition of historical dialogue
around the Categories, and that the organization of the ‘historical
dialogue’ (as described in the concluding remarks of the next chapter)
provides some insight into the stratigraphy of Simplicius’ sources.
However, the stratigraphy is not always or necessarily chronological.
For ease of reference, I number the texts below.

Table of responses: Lucius, Nicostratus, and their respondents

Ref. In Cat. Primary 2nd Respondent 2nd respondents

aporist aporists

T1 1,19 Not applicable: Simplicius’ description of Lucius and Nicostratus’

tactics.
T2 21,2–25,9 ˇƒ — æd N. ‘they’ Porphyry Iamblichus
T3 26,21–28,1 ˇƒ — æd N. ‘they’ ‘Some
[commentators]’
T4 29,23–30,5 ˇƒ — æd N. Porphyry
T5 30,16–22 Nicostratus Atticus Porphyry
T6 48,1–50,3 ˇƒ — æd L. Porphyry Boethus
T7 58,14–28 ˇƒ — æd N. [ˇƒ — æd Boethus Simplicius
L.]
T8 62,28–65,14 ˇƒ — æd Simplicius Boethus
L. & N.
T9 73,15–76,16 Plotinus ˇƒ — æd Simplicius Boethus, Porphyry
L. & N.
T10 125,16 ˇƒ — æd L. Simplicius
T11 127,11–128,10 ˇƒ — æd Plotinus Simplicius Porphyry,
L. & N. Iamblichus, Plotinus
T12 156,16 ˇƒ — æd L. Simplicius Porphyry
T13 231,19–23 Nicostratus Iamblichus
T14 257,31–6 ˇƒ — æd N. Iamblichus?
T15 268,19–32 ˇƒ — æd N. Iamblichus
T16 368,1–371,27 Plotinus Nicostratus Simplicius Iamblichus
T17 381,3–32 ‘Peripatetikoi’ Iamblichus
T18 385,9–411,28 Nicostratus Simplicius
T24 (passim)
T25 414,23–415,15 Nicostratus ˇƒ — æd e —æçæØ
T26 428,3–430,4 Nicostratus Simplicius

In each instance, Simplicius tends to open the discussion with a

problem posed by the ‘Lucians’ or by Nicostratus’ school, and then
describe, often in a generalized dialogue form, the positions taken by
various historical commentators with respect to this problem. In
some cases the dialogue is represented as a single question and single
 Lucius and Nicostratus 115
answer. (For a survey of Boethus’ role in this pattern of dialogue, see
also ch. 6).
As the table shows, explicit references to the Lucius source appear
to trail off after the discussion of substance (up to 155,30—and 156,16
is also a reference to quality, not to relation); this admittedly scanty
evidence might imply that Lucius was not conducting a line-by-line
commentary on the whole text, but may have been especially inter-
ested in the category of substance. Nicostratus’ concerns seem more
wide-reaching. But this may represent Simplicius’ (or his sources’)
own selection from their predecessors, rather than the focus of the
Lucius and Nicostratus sources themselves.

Respondents to the Nicostratans

The core set of problems credited to the Nicostratans tends to be
handled directly by Porphyry (T2, T4–T6 in the numbering of the
table in the preceding section), and in one case by ‘Porphyry’s
followers’ (T25). For example (T2):
Nicostratus and his followers raise the following problem: ‘Why is it
that Aristotle, having proposed to speak about the categories, does not
teach us about them right away, but rather about the other things; viz.
homonyms, synonyms, and paronyms?
Porphyry gave a fine reply to these objectors, saying that in the case of
almost every theoretical field of study, some things are written before-
hand, which tend to clarify what is to follow, as for instance, definitions,
axioms and postulates in geometry (21,2–7, tr. Chase 2003).24
As is often pointed out, it seems likely that Simplicius’ source for
these Nicostratan problems is Porphyry’s commentary Ad Gedalium
and Iamblichus’ commentary. Moreover, Porphyry appears to be
Simplicius’ source for the sequence of dialogue with other historical
figures such as Boethus (see for example T9) and Atticus, who, in
Simplicius’ view, develops and improves upon Nicostratus’ IæÆ
that homonyms are synonyms (see T5 and Anton 1969:16).

24
æFØ ƒ  æd e ˝ØŒæÆ,    æd H ŒÆŪæØH N E
æŁ  P  æd ÆPH PŁ , Iººa  æd H ¼ººø Ø Œ Ø H  › ø ø
ŒÆd ıø ø ŒÆd Ææø ø. æe o ŒÆºH Å  › —æçæØ , ‹Ø ŒÆŁ’
ŒÅ å e Ł øæÅØŒc æƪ Æ Æ æªæç Æ ØÆ æe Æç ØÆ  Æ
H B , ‰ K ª ø æÆ
fi ƒ ‹æØ ŒÆd a IØ ÆÆ ŒÆd a ÆN ÆÆ· (ed. Kalbfleisch
1907).
116 Questions and Responses
Simplicius highlights the reliance of certain passages in Plotinus
Enneads 6.1 upon Nicostratus, particularly later in the commentary.
In several long back-and-forth exchanges (e.g. T9, T11, and T16)
Simplicius sketches a complex stratigraphy of question and answer,
from which one sometimes derives the peculiarly atemporal impres-
sion that Nicostratus developed certain problems raised by Plotinus,
until Boethus helped to solve them. Particularly where Plotinus is
involved, however, and in the discussion of the later categories,
Simplicius becomes less concerned with the discussion of ‘historical’
figures, and strikingly begins to respond directly to ‘Nicostratus’
rather than the previously favoured locution ƒ  æd e ˝ØŒæÆ.
In these later passages where the new language predominates, Por-
phyry appears less prominent: Simplicius frequently cites Iamblichus
as his source for solutions (T13–T15, T17), and also introduces the
previously unused locution ƒ  æd e —æçæØ (T25). In these
later parts of the commentary, it seems probable that Porphyry’s Ad
Gedalium is being read through the lens of Iamblichus’ commentary
(perhaps more than it was earlier), as Simplicius’ own introductory
remarks (in Cat. 3,1–4) suggest.25
After these Iamblichus citations are introduced, Simplicius grad-
ually ceases to cite explicit sources for the solutions to Nicostratus’
problems. Instead, Simplicius offers solutions to the final Nicostratan
problems in his own voice—but these solutions likely belong to
Iamblichus as well, as described by Simplicius in the proem to his
commentary:
Now, I have read some of the aforementioned writings, and, following
Iamblichus as carefully as possible, I wrote them down, often even using
the philosopher’s very words. [ . . . ] I wished to reduce this man’s lofty
spirit, inaccessible to the common people, until it was more clear and
commensurate [with the common understanding] (Simplic. in Cat.
3,2–10, tr. Chase 2003).26

25
Simplicius’ account of his own practice, given in the opening pages of his
commentary (1–3), suggests that he recognized an historical progression of commen-
tators including Themistius, Alexander, Herminus, Maximus, Boethus, Lucius,
Nicostratus, Plotinus, Porphyry, Dexippus, and Iamblichus, and that his primary
(direct) source was the commentary of Iamblichus, supplemented with ‘some’ (3,1)
direct reading of the earlier sources that Iamblichus cited.
26
Kªg ªaæ Kıå b ŒÆ ØØ H NæÅ ø ıªªæ ÆØ, KØ º æ b
‰ x   q E  Æ ºåı ÆæÆŒºıŁH I ªæÆł Å, ŒÆd ÆPB fi ººÆåF B
fi
º Ø F çغçı åæÅ  . › b Œe q Ø B IªæÆçB – Æ b ‰ i
 Lucius and Nicostratus 117
A good example of this system of questions and responses is
368,1–371-27 = T16. Nicostratus notes that it is unclear which, if
any, among the senses of ‘having’ at Cat. 15b17–30 answer to the
category. He goes on (368,26–369,14) to break down the senses in
detail, assigning the first (state, condition) to Quality; the second
(size) to Quantity; the third, fourth, and seventh (things-around, as
on a part, as a possession) to Relation; the fifth (as a part) to Relation
or Substance; and rejecting the eighth (as husband to wife),27 and
concluding that none of these answers to the category. Simplicius
now brings Iamblichus on stage, who retorts that the force of the
category is concealed in the third sense (things-around) and the
fourth sense (on a part), and that Nicostratus is simply insufficiently
graced with IªåØÆ to notice this (370,5).
The subsequent introduction of Plotinus (Enn. 6.1.23) at 370,11—
portrayed as if he is jousting with a contemporary Iamblichus across
the room, in a classic sample of ‘Simplician dialogue’—likely derives
from Iamblichus’ own commentary, and perhaps, through it, from
Porphyry. Thus it is inserted here after Iamblichus’ initial response to
Nicostratus and before the longer discussion of Metaphysics 5.23 on
the senses of having (largely a paraphrase of 1023a8 and following).
This entire back-and-forth, including a report of Boethus, is then
wrapped up at 374,7 where Simplicius mentions that Iamblichus
has just completed c F å Ø  庪Æ, and will now proceed
to deliver a higher criticism, B  æA  æd ÆPF Ł øæÆ
IغÆ  . Iamblichus’ search for an ‘intelligible’ account of the
category, then, begins from a critical engagement with Nicostratus
and Plotinus, both of whom are concerned with the problem whether
the Categories’ analysis of substance attempts to account for both
intelligible and sensible substance (76,14–17; cf. Plotinus 6.1.2).28
The ‘wrap-up’ at 374,7 strongly suggests that Simplicius is drawing
on Iamblichus throughout the preceding pages, including the report

ıÆ Å KŒ F ªæç Ø IŒæØ æÆ (5) H NæÅ ø ŒÆÆÅØ æºÆ E, – Æ
b e łÅºe F F I æe ŒÆd E ººE ¼Æ Kd e Æç æ  ŒÆd
ı æ æ ŒÆƪƪ E, KŒ æø b ŒÆd e ºf ºBŁ H ºı Ø H
ıªªæÆ ø K’ ºÆ ›øF ı EºÆØ, På oø ‰ › çغç Æ
ıæØÆe N KºåØ, Iººa a F Å b ‰ ıÆe H IƪŒÆø ÆæƺØ E
(ed. Kalbfleisch 1907).
27
Nicostratus does not appear to comment on the sixth sense (‰ K Iªª ø fi ).
28
van den Berg 2008 offers an excellent overview of the later Neoplatonist debate
from Plotinus to Proclus.
118 Questions and Responses
and criticism of Nicostratus and the dialogue with Plotinus. Indeed,
I note a problematic section (371,23–4) where Simplicius has been
interpreted from the context29 to cite Plotinus via çÅ, although
these words do not occur in our text of Plotinus 6.1.23 (Iºº’ ¼ ,
çÅ, I ŒØ Çø e ºª Ø å ŁÆØ ªıÆEŒÆ ÅŁ Z). It seems
plausible that çÅ here could alternatively represent Iamblichus,
presenting a point of view (not necessarily Plotinus’ own) in the voice
of a rhetorical interlocutor.30

Respondents to the Lucians

The situation is rather different with the first portion of the commen-
tary, throughout Simplicius’ representation of respondents to the
‘school of Lucius’. Boethus figures heavily here, as in T6, T7 (impli-
citly; see discussion below), T8, and T12.
An especially important passage for the relationship between
Lucius and Boethus is 73,15–76,16 = T9, and I would like to begin
here. In this passage, Lucius (followed by Nicostratus and Plotinus, if
the citation at 73,27–8 is read chronologically) question whether the
ten Aristotelian categories can describe intelligible as well as sensible
beings. It appears that Lucius or his school may have been the first to
raise this problem (which was amplified by Nicostratus and Plotinus).
Simplicius, Boethus, and Porphyry are among the respondents, with
Porphyry correcting Boethus.
If someone wonders whether Aristotle here divided and enumerated
sensible and generated things alone, or whether he did so with all
beings, whatever their mode of existence; and, in general, whether
the intelligible genera are different from the sensible; or whether some
are the same, and others different. After all, if they are different, the
intelligible genera have been completely omitted. If, however, they

29
See Gaskin’s translation (2000), 106 and note 550.
30
There is also an argument advanced by P. Henry (1987: 123 and following) that
Porphyry had access to oral lectures of Plotinus, and that passages that appear to be
Plotinian, yet are not presented using the exact text of the Enneads (such as 73,15 and
following) could represent these lectures. Michael Chase (2003: 152) suspects that the
mediation of Porphyry’s Ad Gedalium commentary, and of Iamblichus’ commentary,
is sufficient to explain the difference of text. Certainly, on my view in the case of the
present passage, Iamblichus may have (nearly) paraphrased the arguments of Plotinus
without clearly endorsing or rejecting them, before providing his own ‘intellective’
interpretation.
 Lucius and Nicostratus 119
are identical, sensible things will be synonymous with intelligible things.
But how could there be commonality (ŒØøÆ) in the same substance
among things in which there is something ‘prior’ and ‘posterior’, and
one of which is a model (Ææ ت Æ), the other an image? If the ten
categories are said homonymously of the intelligibles, they will not be
the same, since they will have in common only the same name; but there
will be more genera, since the intelligible things have not been included.
Moreover, how can it be other than incredible that ‘being-affected’ (e
å Ø) and the ‘relatives’ (a æ Ø) are in the intelligibles, when the
latter are immutable; or that which resembles an offshoot (ÆæÆçı Ø)
among things which stand primarily in self-identity? If, however, some
are common to intelligibles and sensibles, and some are peculiar (N Æ),
their articulation ( ØæŁæøØ ) has been omitted. Such, then, are the
problems raised by the most divine Plotinus, as well as by the followers
of Lucius and Nicostratus.31 (73,15–28, tr. Chase 2003)
This launches a long doxographical passage that mingles the concerns
raised by Plotinus, Nicostratus, and Lucius with several layers of
responses that must date from different points in the commentary
tradition. (The question whether substance can be a single genus is
reiterated at 76,13–15, now attributed only to Plotinus and Nicostra-
tus; but the earlier passage above seems to suggest that Lucius also
offered a similar puzzle, which Nicostratus and Plotinus refined).
Simplicius subsequently offers a response developed by Boethus
(78,4–20, subsequently rebutted by Porphyry beginning at 78,20),
which chronologically seems certain to be directed at a reader before
Nicostratus (most likely Lucius, who has already been named at the
end of 73,15–28, my T9 above):32

31
¯N  Ø Iæ E,  æ a ÆNŁÅa Æ ŒÆd ª Åa Ø Eº  K (15) Ø ›
æغŠŒÆd IÅæØŁ Æ j Æ a ›øF ZÆ, ŒÆd ‹ºø  æ ¼ººÆ Kd
a Åa ªÅ j a ÆNŁÅ, j Øa b a ÆP, Øa b  æÆ· ŒÆd ªaæ N b ¼ººÆ,
Ææƺº ØÆØ fiÅ KŒ EÆ· N b a ÆP, ı ı Æ ÆØ a ÆNŁÅa E ÅE ·
ŒÆd H ÆØ ŒØøÆ B ÆPB PÆ , K x e æ æ Ø ŒÆd e o æ, ŒÆd
e b (20) Ææ ت Æ, e b NŒ ; N b › ø ø ºªÆØ Kd H ÅH ƃ ŒÆ
ŒÆŪæÆØ, PŒ ÆØ Æƒ ÆPÆ, Y æ O Æ ı F ÆPF ŒØøFØ, Iººa
º ø ÆØ a ªÅ, P  æغÅçŁø H ÅH. Ø b H PŒ IŁÆ K
KŒ Ø IæØ sØ r ÆØ e å Ø ŒÆd a æ Ø, ÆæÆçı Ø KØŒÆ K E
ŒÆa a ÆPa æŪı ø HØ; N b a (25)  KØ ŒØa ÅE ŒÆd
ÆNŁÅE , a b Y ØÆ, Ææƺº ØÆØ  ø ØæŁæøØ . ÆFÆ b s ŒÆd ›
Ł ØÆ —ºøE Iæ E ŒÆd ƒ  æd e ¸ŒØ ŒÆd ˝ØŒæÆ (ed.
Kalbfleisch 1907).
32
As Plotinus and Nicostratus (with Praechter’s identification) lived too late for
Boethus to have responded to their criticisms directly; see ch. 6 below. Chiaradonna
120 Questions and Responses
However, Boethus considers these questions redundant here, since <he
believes that> the discussion is not concerned with intelligible sub-
stance. Rather, he says, one should have raised the additional problem
that in other works <Aristotle> divides substance into three: he said that
substance is spoken of in one way as matter, in another as form, and in
yet another as the composite, but that here he posits substance as one
category. What is this substance, and how will <Aristotle> arrange the
three <substances> under it, given that they are not called substances in
virtue of one account (ºª )? In answer to these problems Boethus
claims that the account of primary substance fits both matter and the
composite... matter and the composite will be subsumed under the
category of substance, but the form will be outside the <category of>
substance, and will fall under a different category, either quality or
quantity or another one.33
This passage, I think, also illustrates the tone of Lucius’ IæÆØ, in
contrast to Nicostratus’: where the later problems raised by Nicos-
tratus alone (without evidence of a Lucian predecessor) often focus on
the rhetoric and structure of the treatise, as discussed above, Lucius’
concerns tend to be more direct and doctrinal, specifically focused on
the nature of substance in the Categories.
The associates of Lucius appear at 48,1–50,3 = T6 to argue that (A)
‘the things which complete substance’ (a ı ºÅæøØŒa B PÆ )
cannot be ‘in a substrate’, certainly not without being ‘parts’, and
moreover that (B) Aristotle’s definition would make even Socrates
appear to be ‘in a substrate’. Porphyry answers the first problem (with
some additional commentary by Simplicius himself). Here, for refer-
ence, is that exchange:
[T6 (A). 48,1–49,30] Lucius and his circle, however, also raise the
following objection against the fact that what is in a substrate is said

2009: 104 develops an appealing case for Eudorus as the target of Boethus’ reply. See
also Barnes 2005a.
33
 ˇ Ø BÅŁ ÆFÆ b Æ溌 Ø KÆFŁÆ a ÇÅ ÆÆ º ÆØ· c ªaæ
r ÆØ  æd B ÅB PÆ e ºª· Aºº b  Ø, çÅ, (5) æÆæ E ‹Ø K
¼ººØ c PÆ Ø º  N æ E ¼ººø b c oºÅ, ¼ººø b e r  , ¼ººø b
e ıÆ ç æ PÆ ºª ŁÆØ r  , KÆFŁÆ b Æ Ł ÆØ ŒÆŪæÆ c
PÆ. Æ s ÆÅ, ŒÆd H ÆPB fi a æ E  Ø a c ŒÆŁ’ Æ ºª
º ª Æ ; IÆH b æe ÆFÆ › BÅŁ e B æ Å PÆ ºª ŒÆd B fi
(10) oºfiÅ ŒÆd fiH ıŁø
fi KçÆæ  Ø çÅ . . . Iºº’ oø , çÅd › BÅŁ ,  oºÅ
ŒÆd e Ł  ÆåŁÆØ B fi B PÆ ŒÆŪæÆ fi , e b r  B b PÆ
KŒe ÆØ, ’ ¼ººÅ b   EÆØ ŒÆŪæÆ, XØ c ØÅÆ j ÅÆ j ¼ººÅ
Ø (ed. Kalbfleisch 1907).
 Lucius and Nicostratus 121
to be ‘not as a part’ [at Cat. 1a24–1b3]. ‘For if,’ they say, ‘we say that
things which complete a substance are parts of that substance, and that
which simply completes the being of a sensible body is colour, figure,
magnitude, and simply quality and quantity (since there could be no
colourless or figureless body), and of this particular body, snow, for
instance, whiteness and coldness [sc. complete the substance], then one
of two things is necessary: either not to say that these things are in a
substrate, or [to say that] it was not correct to deny of things in a
substrate that they are like parts. How, moreover, is it possible for
completers (a ı ºÅæøØŒ), in general, to be said to be in a sub-
strate? For Socrates’ shape ( æç) is not in Socrates as its substrate;
rather, if anything, it would be those things which enter into already
complete things from outside which would be in them as their
substrate.’
Porphyry solves this difficulty in the following way: ‘There are,’ he
says, ‘two kinds of substrate . . . Qualityless matter, which Aristotle calls
“potential body”, is the first meaning of “substrate”, and the second is
that which comes into existence as either a commonly qualified thing or
as something individually qualified . . . Therefore,’ he says, ‘many of the
things which inhere are in a substrate with regard to the first substrate;
for instance, all colour and all figure and all quality are in prime matter
as their substrate, not as parts of it and incapable of existing apart from
it. In the case of the second substrate, however, not all colour nor all
quality is in a substrate, but [they are so only] when they are not
completers (ı ºÅæøØŒÆ) of substance.’34 (tr. Chase 2003)

34
æFØ b ƒ  æd e ¸ŒØ ŒÆd F æe e c ‰ æ ºª ŁÆØ e K
Œ Ø ø fi . “ N ªaæ a ı ºÅæøØŒa B PÆ , çÆ, æÅ ÆPB r ÆØ ºª ,
ı ºÅæE b ±ºH b F  Æ F ÆNŁÅF e r ÆØ åæH Æ åB Æ ª Ł ŒÆd
±ºH ØÅ ŒÆd Å (¼åæı ªaæ ŒÆd IåÅ Ø PŒ i YÅ H Æ), F b
F  Æ (5) x  åØ º ıŒÅ ŒÆd łıåæÅ , IªŒÅ ıE Ł æ, j c ºª Ø
ÆFÆ K Œ Ø ø fi j c ŒÆºH IçŒ ŁÆØ H K Œ Ø ø fi e c ‰ æÅ
r ÆØ. H b ŒÆd K Œ Ø ø fi ‹ºø a ı ºÅæøØŒa ıÆe ºª ŁÆØ; P ªaæ 
øŒæı æçc K Œ Ø ø fi Kd fiH øŒæ Ø, Iºº’ Y æ ¼æÆ, a E X Å
 º Ø øŁ  K ØØÆ, ÆFÆ i YÅ K (10) Œ Ø Ø ÆPE ”.
ÆÅ c c IæÆ ºø › —æçæØ “ Ø, çÅ, Kd e Œ  , P
 ŒÆa f Ie B A , Iººa ŒÆd ŒÆa f æ ıæı · X  ªaæ ¼Ø
oºÅ, m ı Ø ŒÆº E › æغŠ, æH KØ F Œ Ø ı Å ÆØ , ŒÆd
 æ, n ŒØH Øe j N ø çÆÆØ· Œ   ªaæ ŒÆd › åƺŒ KØ (15)
ŒÆd › øŒæÅ E KتØ Ø j ŒÆŪæı Ø ŒÆ’ ÆPH. ººa s, çÅ,
H KªªØ ø ‰ b æe e æH Œ   K Œ Ø ø fi K, x  A
åæH Æ ŒÆd A åB Æ ŒÆd AÆ ØÅ K Œ Ø fiÅ Kd B fi æ fiÅ oºfiÅ, På ‰
æÅ ÆPB ZÆ ŒÆd I ÆÆ åøæd ÆPB r ÆØ· Kd b F ıæı Œ Ø ı P
A åæH Æ P b AÆ (20) ØÅ K Œ Ø ø fi , Iºº’ ‹Æ c ı ºÅæøØŒÆ NØ
B PÆ . (ed. Kalbfleisch 1907).
122 Questions and Responses
But Boethus is cited as directly answering the second problem. Here,
then, is another fairly clear-cut case of Boethus replying to a ‘Lucian’
problem:
[T6 (B). 49,31–50,9] ‘But why,’ they say, ‘will not individual substances,
such as Socrates and Plato, be subsumed under the account (ºª ) of
the things “in a substrate”, and be accidents, since Socrates is also “in
something” —for he is in time and space—and not as a part, and it is
impossible for him to exist apart from time and space? For even if he
were to change from this particular place, he will still be in some other,
as was said in the case of fragrance.’
Boethus thought he solved the difficulty based on place, when he said
that things in motion are, in general, never in the place in which they
were, for this was proved in the On Motion. By the same reasoning
(ºª ), however, neither would they be in partial time. For since time
flows constantly, it is always other; so that if anything, they are in
universal time. Boethus, however, provides a solution for this, too. In
the first place, he says, universals are not even in existence (K
 Ø) according to Aristotle, and even if they were, they are not
a determinate thing (Ø). Yet Aristotle said ‘in some thing’. Therefore,
that which is ‘in something’ cannot be in a universal.35 (tr. Chase 2003)
The followers of Lucius may be making this case in order to prove
that qualities are ‘substantial’ in a way that, on his view (like
Eudorus’), Aristotle’s argument would not allow.36
At T8, Simplicius begins a comprehensive description of commen-
tators who ‘opposed the division’ of the ten categories. He describes
the Nicostratans among those who declared the division deficient,
and continues this view to the followers of Lucius. They, according to
Simplicius, held that Aristotle wrongly omitted conjunctions, articles,

35
ººa H , çÆ, Påd ŒÆd ƃ ¼ Ø PÆØ, x  øŒæÅ ŒÆd —ºø, fiH
ºªø fi H K Œ Ø ø fi ÆåŁÆØ ŒÆd ÆØ ı  ÅŒÆ, Y æ ŒÆd  Ø
KØ › øŒæÅ (K ø fi ªaæ ŒÆd åæø
fi ) ŒÆd På ‰ æ ŒÆd I Æ åøæd r ÆØ
ı ŒÆd åæı; Œi ªaæ F [p. 50] ÆB fi F ı, ø K ¼ººøfi ÆØ, ‰
Kd B Pø Æ Kºª .
Iººa e b Ie F ı º Ø › BÅŁ    Ng a ŒØ Æ c r ÆØ ‹ºø
K fiz q ø fi · F ªaæ K E  æd ŒØ ø EåŁÆØ. fiH b ÆPfiH ºªø
fi P b K
åæø fi YÅ i æØŒfiH· Þ ªaæ ı åH F åæı, ¼ºº I d ŒÆd ¼ºº K, u ,
Y æ ¼æÆ, K fiH ŒÆŁºı (5) åæø fi K. Iººa ŒÆd F ºø › BÅŁ æH b
P b r ÆØ e ŒÆŁºı K  Ø ŒÆa e æغŠçÅ, N b ŒÆd YÅ, P d
r ÆØ· › b æغŠ ØØ r  · u P ÆÆØ e K Ød k K fiH ŒÆŁºı r ÆØ.
(ed. Kalbfleisch 1907).
36
cf. 156,16 = T12 in the table above, and my discussion of Eudorus’ possible
argument in favour of intelligible qualities and quantities perceived absolutely, ch. 3.
 Lucius and Nicostratus 123
negations, privations, moods of verbs, the monad, and the point
(64,13–65,14). (I believe that the linguistic concerns in 64,18–20
may be wrongly ascribed to the Lucians, for reasons I argue in ch. 5
in treating Athenodorus and Cornutus;37 however, 65,13–14 is,
I believe, correctly assigned.) Simplicius appears to respond directly
to the linguistic omissions from 64,18–65,3, presumably relying on
Iamblichus and Porphyry. But for the objection at 65,13–14, namely
the complaint that the categories will not embrace the monad and the
point, and similar (presumably ‘intelligible’) entities, we have a reply
by Alexander and by Boethus; Boethus’ response offers a distinctive
solution that accords some substantiality to these mathematical and
geometrical entities, including an argument that these entities may be
both substance and quantity.
‘But what about the one,’ they say, ‘and the monad and the point? How
can they not fall outside the categories? They are not, as one might
think, quantified, for they are neither continuous—since they are with-
out parts—nor are they discrete. But it is fitting that everything quan-
tified should be either continuous or discrete, and if discrete, then either
odd or even.’38
The answer is, as Alexander also holds, that they will be placed
among the Relative, both as principle (Iæå) of numbers and as meas-
ure. If, however, number is twofold—one incorporeal, the other
corporeal—then, as Boethus would say, the monad will also be twofold:
one which is substance (PÆ), and is in intelligible number—Aristotle

37
There are some oddities in this passage, and, in ch. 5, I propose that it may be
wrongly ascribed to the followers of Lucius. While some of the discussion in that
chapter is specific to Athenodorus and Cornutus, two points may be briefly made
here. First, Lucius is unusually named here apart from Nicostratus, and most
strangely, he is named immediately after Nicostratus with a differing view. Simplicius
does not elsewhere set ‘Lucius’ and ‘Nicostratus’ against one another in this way, nor
in this order. Second, just after the ascription to Lucius, but before the Porphyrian
answer that is elsewhere associated with the Stoics, our text of Simplicius has a lacuna.
This issue is discussed in further detail in ch. 6.
38
Iººa e , çÆ, ŒÆd  a ŒÆd e Å E H PŒ ø  Ø H
ŒÆŪæØH; h ªaæ  KØ, ‰ ¼ ø fi  Ø · h ªaæ ı å KØ, –
I æb Z, h ØøæØ · A b e j ı åb j ØøæØ  r ÆØ æŒ Ø, ŒÆd
N ØøæØ , j  æØe j ¼æØ. j ŒÆd ‰ Iæåc H IæØŁ H ŒÆd ‰ æ K fiH
æ Ø ŒÆÆÆåŁ ÆØ, ‰ ŒÆd fi H º  æø fi Œ E. N b Øe › IæØŁ  , › b
I Æ , › b ø ÆØŒ , ÆØ, ‰ ŒÆd
› BÅŁ i çÆÅ, ŒÆd  a Ø,  b PÆ,  K fiH ÅfiH (20) IæØŁ fiH (F
b ŒÆd æغ Ø Œ E),  b æ Ø j . o æ  çÅØ › BÅŁ 
¼ Ø r ÆØ e çÆØ· ‰ ªaæ º ıŒÅ å Ø æe º ıŒ, oø ŒÆd  ıa æe a
· N s KŒ EÆ ¼ çø Ø, ŒÆd ÆFÆ . (ed. Kalbfleisch 1907).
124 Questions and Responses
also thinks that this one exists—and one which is a relative or a
quantified. Later, however, Boethus says that perhaps it is better to
call it a quantified, for as whiteness is to white, so is the dyad to two.
If, therefore, the former are both qualified, then the latter are also
quantified. (65,13–24, tr. Chase 2003)
Here, too, it looks as if Boethus may be defending against a problem
initially posed by Lucius.
Early during the discussion of Quantity, Simplicius replies directly
and in his own voice to the Lucians (125,16 = T10) when they
question why Aristotle transfers body to the category of Quantity
rather than substance. His solution is markedly similar to that
ascribed to Boethus at 65,19–21 = T8: much like the monad and the
point, body is both substance and quantity. It is possible that the
Boethan both-and response implicitly underlies the comparable solu-
tion to T10 as well. Similar issues arise in other discussions of the
category of Quantity.39
I think we can also find less clear-cut but still promising examples
of Boethus replying to a Lucian problem. At 58,14–28 = T7, for
example, the Nicostratans are (A) credited with a rather tendentious
attack on Aristotle’s ‘unnecessary verbiage’ ( ÆÆغªÆ), and ‘they’
(third person plural) are further credited (B) with an objection to the
Aristotelian position that the substrate has all the differentiae of the
predicate: how, they ask, can the rational animal inherit both
‘rational’ and ‘irrational’ (e  ºªØŒe ŒÆd e ¼ºª) from the
genus ‘animal’?
[T7 (A). 58,14–22] Nicostratus and his followers reproach Aristotle
with indulging in useless verbiage in this passage. ‘For who,’ he [sc.
Nicostratus] says, ‘is unaware that knowledge could never differ from
another knowledge by being two-footed or four-footed, as is the case
with “animal”?’ In the first place [Simplicius replies], even if this much

39
In the broader issues regarding Quantity in a very complex passage
(127,11–128–10 = T11; see discussion in Griffin 2009: ch. 4), the Lucians are included
with the Nicostratans and Plotinus as holding that magnitude is not a quantity, but a
‘so much’—because only Number, properly understood, is truly Quantity. They also
wonder why sensible number is not substance, whereas intelligible number is. Their
doctrinal interest seems to follow T9 (and to a lesser extent T8) where the Lucians are
also concerned with the intelligible–sensible distinction and whether Aristotle has
fully grasped its significance—and are particularly interested in his treatment of
‘substantial’ quantities, including ‘Number’ in an almost Pythagorean sense (129,27
= T11) as well as the Monad and the Point (T8) and ‘body’ itself (T10).
 Lucius and Nicostratus 125
is clear, it is not so obvious that ‘there is nothing to prevent the
differentiae of subordinate genera from being the same’ . . . (tr. Chase
2003)40
[T7 (B). 58,23–8] They also object to [Aristotle’s] statement that, in
subordinate genera, ‘however many differentiae there are of the predi-
cate, there will be just as many differentiae of the substrate’ . . . Now
Boethus gave in to this problem, and suggested emending the text as
follows: ‘so that however many differentiae there are of the subject, the
predicate will have the same number.’ (tr. Chase 2003)41
Boethus is said to have ‘given in’ to the latter problem (K f B fi
IæÆfi) by amending the text.42 But it seems chronologically unlikely
for Boethus to have literally ‘given in’ to a problem first discovered by
Nicostratus.43 With the exception of T7, Boethus replies to ‘Lucian’
arguments but does not reply to ‘Nicostratan’ arguments. I believe
that the third person plural at T7B (but not T7A) refers loosely and
implicitly to the Lucians, and not to the Nicostratans. This would not
be unprecedented, as Simplicius elsewhere introduces a position
ascribed to Nicostratus’ school and then continues the same position
to ƒ  æd e ¸ŒØ as if it makes no difference (64,13–20 = T9).
Thus T7B is, like T6B, an example of Boethus responding directly to

40
ˇƒ b  æd e ˝ØŒæÆ ÆÆغªÆ KªŒÆºFØ K Ø ·  ªæ, çÅ,
Iª Ø ‹Ø KØ Å KØ Å PŒ ¼  fiH  Ø Ø ªŒØ j  æ Ø, u æ
e ÇfiH; Iººa æH , Œi F Æç , PŒØ ŒÆd KŒ E › ø Bº, ‹Ø H
ººÅºÆ P b Œøº Ø a ÆPa r ÆØ ØÆçæ ·  Ø b  æd I çE – Æ H
IØŒ Ø ø N E.  ØÆ ›æfi A ‹Ø ƒ  æd e  Eæ E e  æÆ ŒÆd c ººÅºÆ
P ŒÆºH K Æ, (20)  Æ ıÆe r ÆØ ŒÆd c ººÅºÆ ZÆ ç’ £
Iª ŁÆØ ª · oø PŒ q æå Øæ › ºª , Œi a ÆæÆ ª ÆÆ ŒÆºH
ºÅçŁÆ ÆçB e ºª ØB fi (ed. Kalbfleisch 1907).
41
ÆNØHÆØ b ŒÆd <‹Ø> K E ƺººØ ª Ø ‹ÆØ F ŒÆŪæı ı
çÅd ØÆçæÆ, ÆFÆØ ŒÆd F Œ Ø ı ÆØ· Zø ªaæ ƺººø F
Çfi ı ŒÆd F ºªØŒF Çfi ı, (25) K Ø c F Çfi ı ØÆçæÆ NØ   ºªØŒe ŒÆd e
¼ºª, H x   F ºªØŒF Çfi ı e b r ÆØ ºªØŒ, e b ¼ºª; Iºº’ › b
BÅŁ K f B fi IæÆ fi ƪæç Ø Mı c ºØ oø ·(ed. Kalbfleisch 1907).
42
See the discussion under ‘Differentiae’ in my chapter on Boethus, below.
Boethus reverses the order of the terms ‘subject’ and ‘predicate’ at 1b23–4 to give
‘however many differentiae there are of the subject (Œ Ø ı), there will be just as
many differentiae of the predicate (ŒÆŪæı ı)’. Ackrill, in his commentary
(1963:77) adopts a similar approach.
43
First, Boethus’ birth is (as securely as possible) dated to the sixties bce, and we
have discussed fairly firm, if conjectural, reasons to date Nicostratus’ floruit to the
second century ce. Moreover, there are other instances in Simplicius, such as
29,23–30,5 = T4, which seem to make it clear that Nicostratus cannot have been
active (or, at least, active and widely known) before Boethus or contemporary with
him. There Nicostratus criticizes a reading at 1a1–2 that Boethus clearly did not know.
126 Questions and Responses
the ‘Lucians’; it is also an example of Simplicius roughly conflating
the Lucian and Nicostratan positions (cf. Moraux 1984: 528–63) in
such a way that Lucius appears to underlie Nicostratus.
The final problem explicitly raised by the ‘Lucians’ involves the
substantiality of Quality (156,16–25 = T12). Much as the Lucians
were previously concerned with the substantiality of quantities, they
are here concerned to demonstrate that Quality is closer to Substance
than is the Relative:
And it has greater affinity with Substance, as the followers of Lucius also
object, for [they say that] it is more apt to reveal Socrates from his snub
nose, protruding eyes and pot belly—which are qualities—than from his
being on the right hand, his being a friend and his other features which
are relative. Besides, they say, if things that are spoken of are divided
into two, —into what is per se and what is relative to something else—
when [Aristotle] started to talk about what is per se, which includes
Substance and Quantity, he ought to have added Quality and in this way
to have proceeded to Relatives. If according to Aristotle himself that
which is relative is like an offshoot and supervenes on what is conceived
of as per se, as something posterior, how could [Relatives] fail to be after
Quality? (tr. Fleet 2002).44
Simplicius responds directly and at length to this concern, drawing on
Porphyry’s defence of the Aristotelian order, although that may in
turn derive from Boethus.

Summary of Treatment by Respondents

In those passages where the ‘Lucians’ are directly cited, they tend to
underlie the positions of Nicostratus (and often Plotinus), particu-
larly in complex passages such as T11. When Boethus is cited early in
the commentary, he tends to respond directly to issues raised by the

44
ήd B
fi PÆ
fi b NŒ Ø æ, ‰ ŒÆd ƒ  æd e ¸ŒØ KªŒÆºFØ· NŒ Ø æ
ªaæ e øŒæŠźHÆØ Ie F Ø F ŒÆd F KçŁº ı ŒÆd æªæ , – æ
Kd Ø, j Ie F ØF ŒÆd çºı ŒÆd H ¼ººø H æ Ø. ¼ººø , çÆ, N
 ØÆØæı ø H º ª ø, Y  e ŒÆŁ’ (20) Æe ŒÆd N e æe  æ,
Iæ   æd H ŒÆŁ’ Æe ºª Ø, K x  PÆ ŒÆd e ,  Ø ŒÆd e Øe
æŁÆ oø Kd a æ Ø ÆBÆØ. N b ŒÆd ŒÆ’ ÆPe e æغÅ
ÆæÆçı Ø ØŒ  a æ Ø ŒÆd E ŒÆŁ’ Æa ı Ø ‰ o æÆ Kت ÆØ,
H PŒ i YÅ a e Ø; (ed. Kalbfleisch 1907).
 Lucius and Nicostratus 127
Lucians; on the occasion when he appears to be made to reply to
Nicostratus (T7), it is likely that the Lucians still underlie the argu-
ment and it is to them that he is responding. (Simplicius is not
especially concerned to represent the argument in chronological
order, as the organization of for example Herminus, Athenodorus,
Cornutus, Lucius, and Nicostratus [T8] makes clear.) It is later in the
commentary when other figures, especially Iamblichus, respond dir-
ectly to ‘Nicostratan’ points taken over by Plotinus.
The Lucians are concerned with metaphysical issues of Platonic
doctrine. Their concerns are often shared by Plotinus (and by exten-
sion, in some passages, by the Nicostratans). They worry about the
substantiality of attributes like quantities and qualities; the distinction
of intelligible from sensible reality; and whether the Categories’ sys-
tem is capable of describing these distinctions and making metaphys-
ical room for entities such as intelligible, substantial Number. These
concerns are shared with Eudorus specifically, and more generally
with Neo-Pythagoreanism and Middle Platonism. We might specu-
late that Lucius was an early proponent of Eudorus’ views, and that
Lucius proceeded to apply Eudorus’ critique directly to the Aristotel-
ian treatise in a targeted, critical treatise. Perhaps that treatise elicited
a direct reply from Boethus. The responses that Boethus offered (as
Simplicius represents him) suggest that even Boethus was prepared to
grant a kind of substantiality to certain quantities in response to
Lucius (see ch. 6 for discussion). Moraux (1973: 155) suggests that
Boethus is influenced by Speusippus, and therefore ‘an die substan-
tielle Realität von intelligiblen Zahlen glaubte’; although Tarán (1981)
criticizes this interpretation, Boethus’ thinking about quantity does
seem to have been formulated in response to such ‘Pythagoreanizing’
and ‘Platonizing’ difficulties about the Categories. Thus I would sug-
gest that Boethus responded to ‘Lucian’, doctrinal concerns raised
against the interpretation of the Categories favoured, perhaps, by
Andronicus and Eudorus. The Nicostratans, by contrast, have a
broader set of textual, rhetorical, and logical issues to discuss, which
seem to have interested Atticus (30,16—22 = T5) as well as Porphyry
and Iamblichus. However, they often share the more specific interests
of the Lucians (e.g. T8, T9, T11). ‘Nicostratan’ ideas, particularly
when they appear to overlay ‘Lucian’ questions, anticipate Plotinus
(T9, T11, T16).
Here I would like to stress Lucius’ role in developing criticisms that
presuppose an ontological reading of the Categories (similar to that
128 Questions and Responses
adopted by Eudorus and Pseudo-Archytas), especially reflecting its
limitations in addressing intelligible (not perceptible) beings such as
the monad and the point. I also develop in more detail below (ch. 6)
the suggestion that a commentary by Boethus may have been the
original source for many of Lucius’ IæÆØ.
 5

Stoic Critique: Athenodorus and Cornutus

Seleucus. Well, then, what is the reason that impelled the

philosophers of old to engage in disputes of every kind with
each other on the subject of this Aristotelian writing which we
call the Categories? For as far as I can see, neither have more
numerous controversies occurred about any other topic, nor
have greater contests been stirred up, not only by Stoics and
Platonists trying to undermine these Aristotelian categories, but
even among the Peripatetics with each other. . . .

Dexippus. Because . . . the subject (
Œ) of this book con-

cerns the primary and simple expressions (º Ø) <and the
realities> they signify. So since ºª is useful to all branches
of philosophy, and the first principles of this are simple expres-
sions and their objects of reference, it is natural that much
controversy has arisen . . . .1
Dexippus in Cat. 5,16–22 (tr. Dillon 1990, lightly modified)

1
Æ. s  q ¼æÆ e ØB
Æ f ƺÆØf çغ
çı ØŒ ºÆ ŒÆd
Æ Æa K
åÅŒ ÆØ æe Iºººı æØÆ  æd  ı F æØ
 ºØŒF

ıªªæÆ , n c ŒÆºF ˚Æ Åªæ Æ;
å e ªaæ ŒÆ Æ ÅŒÆ, ‰ h º ı
I غª ÆØ N  æÆ Ł
Ø ª ª Æ
Ø h  Çı IªH  Œ Œ Å ÆØ P  
E  øœŒE ŒÆd —ºÆ ø ØŒE
ƺ  Ø KØå ØæF
Ø Æ Æ a æØ
 ºı
ŒÆ Ūæ Æ, Iººa ŒÆd ÆP E ª E — æØÆ Å ØŒE æe Æı , E b Aºº
KçØŒ E
ŁÆØ B ØÆ  Æ I æe  غÅç
Ø, E ’ Pæ æ º Ø N Ø a
Ææ’  æø Iæ Æ.
 O Ø, t çغÆŁ
Æ º ıŒ ,  æd H æ ø ŒÆd ±ºH º ø , z Æy ÆØ

ÅÆ ı
Ø,
ΠK
Ø K  fi ø fiH غ fi ø· K Øc  ı › ºª A
Ø E Y
Ø
B çغ
ç Æ åæ
Ø,  ı b IæåÆd ƃ ±ºÆE º Ø ŒÆd a e  ø

ÅÆØ  Æ, NŒ ø ººc ªª IçØ
 Å
Ø  æd F ŒÆºH j c ŒÆºH
æØ
 ºÅ Ø ØºÅç ÆØ  æd ÆP H (ed. Busse 1888).
130 Stoic Critique
This chapter focuses on criticisms of the Categories by the Stoic
Athenodorus,2 who, I suggest, presupposed that the Categories
belonged to the rhetorical or grammatical subdivision of logical
investigation, and that Aristotle failed in this field by omitting many
parts of speech.3 (I also argue that Athenodorus’ criticism had noth-
ing to do with a contrast between the four Stoic ‘categories’ and
Aristotle’s genera of predication.) Athenodorus’ complaint—like
Lucius’ critique of the Categories as an inadequate treatment of
ontology—may have helped to inspire Boethus of Sidon’s influential
formula in response: the Categories concerns neither words as such
(with Athenodorus) nor beings as such (with Lucius), but rather
words qua significant of beings: thus some words, and some beings,
will naturally fall outside its scope of reference.
In the course of this chapter, I also try to distinguish different
critiques of the Categories introduced by Athenodorus and the later
Stoic Cornutus.4 I argue that we can distinguish a critique of the
coherence of the work’s division (ØÆ æ
Ø) into chapters and themes,
on the one hand, from a different critique of its division (ØÆ æ
Ø) of
the ten genera of predication, although these two criticisms may be
partially conflated in Simplicius’ doxography. The latter critique
belongs to Athenodorus, but I also suggest that Cornutus may have
rejected this line of attack on the Categories, and along the way helped
to lay the groundwork for Porphyry’s later response to Stoic Iæ ÆØ.
The Neoplatonic commentators represent ‘Athenodorus’ and ‘Cor-
nutus’ as closely linked figures. Athenodorus is a Stoic (Porphyry in
Cat. 86,20, cf. Dexippus in Cat. [in chapter heading] 1,9) who criti-
cizes the Categories for failing to analyse its subject (
Œ)

2
For the historical tradition on Athenodorus, I rely chiefly upon the review in
R. Goulet, DPhA (entries 496–8); Moraux 1984: 592–601; B. L. Hijmans, ‘Atheno-
dorus on the Categories and a pun on Athenodorus’, 105–14 in J. Mansfeld and
L.M. de Rijk (eds), Kephalaion: Studies in Greek Philosophy and its Continuation
Offered to Professor C. J. de Vogel (Assen, 1975). For bibliography see DPhA entries
496–8, and especially entry 497 on Calvus.
3
That line of criticism, as I proposed earlier, might suggest that Athenodorus’
criticisms belonged to a pre-Andronican stage in the Categories’ history.
4
For the following discussion of Cornutus as an historical figure, I am indebted to
the entry of Pedro Pablo Fuentes González in R. Goulet, DPhA (entry 190); Moraux
1984; and the overviews of Cornutus’ allegorical practice in Boys-Stones (2007 and in
Boys-Stones 2003b), in G. W. Most, ‘Cornutus and Stoic Allegoresis’ in ANRW II.36.3
(1989), 2014–65 (with bibliography) and in A.A. Long, ‘Stoic Readings of Homer’
= 1996: 71–2. For comprehensive bibliography see Goulet 1994, DPhA entry 190.
 Stoic Critique 131
adequately. Simplicius depicts Cornutus as following (62,25–7) or as
amending (128,7; 129,1) the criticisms of Athenodorus. According to
Porphyry, their followers (ƒ  æd ŁÅ øæ ŒÆd ˚æ F  ) main-
tain that the Categories treats ‘verbal expressions insofar as they are
verbal expressions’ (º ø ŒÆŁe º Ø, in Cat. 59,5–14),5 and that
the text omits many classes of verbal expression and therefore fails to
treat its proper subject comprehensively (P 
Æ a º Ø
 æغÆF
Æ , Simplicius 19,1). Athenodorus in particular is said to
have rejected the ‘number’ of ten categories, either as too great6 or as
too small.7
In at least one passage of Simplicius’ commentary (18,22–19,7),
Athenodorus and Cornutus are also associated with criticisms of the
literary composition of the Categories, referring to the text as a poorly
organized ‘heap’ of speculations (Æ ÆH Ł øæÅ ø
øæ Æ)
that opens with ‘logic’ (meaning the treatment of homonymy, par-
onymy, and syonymy) before executing a series of loose disciplinary
transitions between theology, ethics, and physics. As I suggest below,
it is not entirely clear that this criticism is rightly attributed to both
commentators, as the passage in question is disjointed, and I hope to
suggest some reasons for peeling the two apart. But modern critics
certainly share these misgivings concerning the structural and the-
matic links of the Categories’ first chapter or ‘onymies’ and last five
chapters or ‘post-praedicamenta’ with its core chapters.8 Simplicius
situates this criticism as a failure to recognize the ‘division of chapters
according to their natural joints’ (ŒÆ ’ ¼æŁæÆ ØfiæÅ ÆØ, 18,23). Since
Athenodorus and Cornutus are claiming that the parts of the text
migrate indiscriminately between logic, ethics, physics, and theology,
this imagery may reflect the Stoic application of Plato’s ‘butcher’
analogy (Phaedrus 265E) to the articulation of philosophy as

5
In this respect they were at variance with Boethus of Sidon, whose alternative
account that the skopos of the Categories was ‘significant sounds insofar as they are
significant’ ( æd çø H
ÅÆ ØŒH ±ºH ŒÆŁe
ÅÆ ØŒÆ N
Ø H æƪ ø ,
Porphyry in Cat. 58,5–6) would win the day in the Neoplatonic tradition. As I suggest
below, however, the notion that spoken words are directly significant of realities is
more Stoic than Aristotelian (de Int. 16a3) or Platonic.
6
Hijmans 1975: 108 and following; cf. M. Pohlenz 1984(6), I, 294.
7
Moraux 1984: 588–9 and n. 18.
8
Contemporary critics who comment on the coherence of the chapters include
John Rist (1989a: 94); Frede (1987: 11ff.); Barrington Jones (1972: 117); and Oehler
(1984: 37–119). There is no evidence that the Stoics, like some contemporary critics,
were led to question the authenticity of the text.
132 Stoic Critique
compounded of several disciplines, logic, ethics, and physics (DL 7.40
= SVF 2.41).
The fragments of Athenodorus and Cornutus on the Categories
raise interesting challenges of interpretation, particularly concerning
their motivations for commenting on this treatise in the first place.
L. Annaeus Cornutus is known to have made a speciality out of
linguistics and rhetoric (Porph. 86,21–2; Simplic. 62,25–6), especially
‘semantics’, and to have written a monograph on a related subject.9
Perhaps, then, his commentary upon the Categories emerged natur-
ally from his own philosophical interests. But Cornutus apparently
criticized some points in Athenodorus’ commentary, enough for his
own work’s title to carry the phrase Against Athenodorus (Simplicius
in Cat. 62,25). Athenodorus’ motivation for studying and refuting the
Categories is more obscure. Indeed, it is not clear which among
several Stoic philosophers of that name should be identified with
the Athenodorus of the Neoplatonic commentators.10
Throughout the following discussion, I seek to support several
claims about the commentaries of Athenodorus and Cornutus on
the Categories. I propose that the defence deployed by Porphyry (and
later by Dexippus, who follows Porphyry,11 and Simplicius) against
Athenodorus’ criticism of the Categories was itself constructed within
a basically Stoic theoretical framework for understanding the ‘impos-
ition’ of names on things. Athenodorus, I think, sought to define the
subject matter of the ten categories in terms of the secondary (lin-
guistic) subdivision of Stoic dialectic.12 This then facilitated his
assessment (as it was subsequently interpreted) that the
Πof
the work was linguistic expressions-as-such (º Ø ŒÆŁe º Ø) and
that Aristotle’s analysis fell short of comprehending these. This read-
ing of the Categories as pertinent to grammar or rhetoric is likely to
have preceded the semantical reading of the Categories adopted
by Andronicus and, in a more nuanced form, by Boethus. At

9
See D. Sedley, ‘Stoic Metaphysics at Rome’ (2005), J. Barnes, ‘Aristotle and Stoic
Logic’ (1999), and Peri Hektōn in POxy 3649; see Hülser, Die Fragmente (1987):
1058–63.
10
See Griffin, ‘Which Athenodorus commented on Aristotle’s Categories?’
(2013b).
11
For the close link between Porphyry and Dexippus, see Hadot 1990.
12
See below for discussion and further literature, especially Allen 2005 and Long
2005, with Long and Sedley 1987: 31–8. K. Hülser (1987) collects the Stoic fragments
on dialectic.
 Stoic Critique 133
any rate, this direction of influence may seem slightly more plausible
than the opposite hypothesis, that Athenodorus responded to an
interpretation of the Categories as ‘teasing out’ the semantical relation
between words and beings with a simple complaint that it failed
effectively to carve up words as such, i.e. that it was a poor grammar
textbook. In any case, there is some supporting evidence, I think, for
dating Athenodorus earlier than is usually supposed, perhaps earlier
than Andronicus; but the central historical suggestions that I would
like to develop here are that Athenodorus treats the Categories as a
basically linguistic work, that there is no evidence of a direct response
to this interpretation by Andronicus, and that Boethus of Sidon will
later take some trouble to rebut this view.
Subsequently, a second and distinct Stoicizing source, I will sug-
gest, responded that the ten categories referenced the subject matter
of the primary subdivision of Stoic dialectic, concerning the so-called
sayables or º Œ , then utilized this analysis to build an entirely
different argument against the coherence of the Categories. According
to this interpretation, the work lacked any unifying
Πand the
subject matter of the ten categories could even be viewed as ‘theo-
logical’ and not linguistic at all. Simplicius appears to combine both
arguments at 18,22–19,7, where they can be distinguished based on
the structure and sources of the passage and its use of language (see
below). I speculate that Cornutus is this second Stoicizing source, and
that this difference constitutes the unidentified ‘disagreement’ repre-
sented by the title of his commentary Against Athenodorus and
Aristotle and his occasional objections to Athenodorus’ ‘lexical’
account of certain categories such as the Relative (cf. Simplicius in
Cat. 187,24–34).
Athenodorus’ line of attack, I believe, involved presuming (or
arguing explicitly) that the Categories really occupied itself with a
linguistic and lexical subject and not with the proper province of logic
and dialectic—the º Œ —and then proceeding to critique the Cat-
egories as an inadequate treatment of the former. Cornutus may have
disagreed, and Herminus, the immediate source of Porphyry’s
defence, certainly disagreed (cf. Porph. in Cat. 59,17–19). Herminus,
here apparently following Boethus, responded by redefining the

Πof the Categories to clarify that it did not concern the kinds
of reality (pace Nicostratus and Plotinus) nor º Ø-as-such (pace
Athenodorus), but instead concerned things that are said ( H
º ª ø , 59,22) and their significance ( c Œ
ı NŒ Æ
134 Stoic Critique

ÅÆ
Æ , 59,26–7)—in Stoic terms, neither words nor realities, but
the ‘sayables’ that somehow mediate between them. The views
ascribed to Herminus by Porphyry also build on a narrative about
the ‘double imposition’ of names that can be attributed to the Stoics
and other Hellenistic schools.13
DL 7.43 reports the bipartition of Stoic dialectic into (a) º Œ  and
(b) vocalization (cf. 7.62–3 for the precise association with º Œ ):14
Dialectic [according to the Stoics] falls under two heads: (a) that
concerning signifieds and (b) that concerning vocalizations ( e  æd
H
ÅÆØ  ø ŒÆd B çø B  ).
(a) And the signifieds fall under the following headings: concerning
çÆ Æ
ÆØ and the º Œ  that arise from these, propositions expressed
and their constituent subjects and predicates (ŒÆ ŪæÆ Æ) and
similar terms whether direct or reversed, genera or species, arguments
too, moods, syllogisms, and fallacies . . . .
(b) And the second main head mentioned above as belonging to
dialectic is that of vocalization (çø ), wherein are included written
language and the parts of speech, with a discussion of errors in syntax
and in single words, poetical diction, verbal ambiguities, euphony and
music, and according to some writers chapters on terms, division, and
style (Loeb translation).15

13
In treating the Stoic and Platonic sources of the theory, I draw in particular on
three recent studies in Frede and Inwood (2005): Allen (2005), Long (2005), and
Ebbesen (2005). See also Sedley (1998), Frede (1989:2088–9 and 1987:301–62), and
Boys-Stones (2001).
14
On this bipartition in particular, see Frede 1987: ch. 16 and Die stoische Logik
(Göttingen: Vandenhoeck and Ruprecht, 1974). On the lekta, see S. Bobzien, ‘Stoic
Syllogistic’, OSAP 1996, ‘Logic’ in the Cambridge Companion and in K. Algra,
J. Barnes, J. Mansfeld, and M. Schofield (eds) 1999, The Cambridge History of
Hellenistic Philosophy (CUP). On Stoic logic in general, see also K. Döring and
T. Ebert (eds) 1993, Dialektiker und Stoiker: Zur Logik der Stoa und ihrer Vorläufer
(Stuttgart: Franz Steiner). K. Hülser (1987) collects the Stoic fragments on dialectic
more broadly.
15
c b Øƺ Œ ØŒc ØÆØæ E
ŁÆØ Y e  æd H
ÅÆØ  ø (5) ŒÆd B
çø B  · ŒÆd e b H
ÅÆØ  ø Y e  æd H çÆ Æ
ØH  ŒÆd
H KŒ  ø çØ
Æ ø º Œ H IØø ø ŒÆd ÆP  ºH ŒÆd ŒÆ ŪæÅ ø
ŒÆd H › ø OæŁH ŒÆd  ø ŒÆd ª H ŒÆd NH , › ø b ŒÆd ºªø ŒÆd
æø ŒÆd
ıººªØ
H ŒÆd H Ææa c çø c ŒÆd (10) [p. 44] a æªÆ Æ

çØ
 ø · z r ÆØ ł ı ı ºªı ŒÆd IºÅŁ  Æ ŒÆd Iç
Œ Æ

øæ Æ ŒÆd f › ı  Ø, KººØ E ŒÆd Iæı ŒÆd  æÆ  Æ ŒÆd
KªŒ ŒÆºı ı Œ æÆ Æ ŒÆd h ØÆ ŒÆd Ł æ Ç Æ. ¯r ÆØ b B Øƺ Œ ØŒB
YØ  ŒÆd e æ ØæÅ  (5)  æd ÆP B B çø B, K fiz  Œ ı ÆØ 
KªªæÆ  çø c ŒÆd Æ a F ºªı æÅ, ŒÆd  æd
ºØŒØ
F ŒÆd ÆæÆæØ
F
 Stoic Critique 135
In what follows, I aim to identify and examine the critiques of the
Categories loosely attributed to the Stoics at large, Athenodorus, and
Cornutus, and I attempt to peel at least some of these apart, particu-
larly two criticisms of a ‘division’ (ØÆ æ
Ø) in the treatise (see II
below). I also suggest, tentatively, that it may have been Cornutus
who found some fault with Athenodorus’ interpretation of Aristotle’s
Categories as a poor shot at (b) the second, grammatical class of
dialectic, and that Cornutus himself rightly engaged with the text as
a contribution to (a) the primary class of dialectic as well.

I. LIFE AND HISTORY

Athenodorus
The ‘Athenodorus’ who commented on the Categories is not precisely
identified by Dexippus, Porphyry, or Simplicius. From these sources,
we know that he was a ‘Stoic’ (Porph. in Cat. 86,20) who wrote his
commentary before Cornutus composed a rebuttal to it (Simplicius in
Cat. 62,26). As Cornutus’ lifetime is well documented, this provides a
terminus ante quem for Athenodorus and establishes his activity
before the second half of the first century ce. The tradition offers
several historical candidates for identification with the Athenodorus
described by Porphyry and Simplicius (who name him) and Dexippus
(who does not). We know of multiple Athenodori of the Stoic school,
all of whom appear to have lived prior to L. Annaeus Cornutus, and at
least two of whom hailed from the city of Tarsus. One of the two best-
documented candidates is the Athenodorus called ‘Cordylion’, a
contemporary of Marcus Cato who became librarian at Pergamon
and was accused of bowdlerizing Zeno’s Republic (DL 7.34); and the
other is Athenodorus ‘Calvus’, identified as the son of ‘Sandon’, a
tutor of Augustus and correspondent of Cicero. Strabo (14.5.14)
informs us that these two Athenodori are best distinguished by
their connections with Cato and Augustus, respectively; the suicide
of Cato in 46 bce establishes a terminus ante quem for the life of
Cordylion, while the son of Sandon died in Tarsus at the age of

ŒÆd ØÅ ø ŒÆd IçغØH ŒÆd  æd K ºF çø B ŒÆd  æd ı
ØŒB ŒÆd  æd
‹æø ŒÆ  Ø Æ ŒÆd ØÆØæ
ø ŒÆd º ø (ed. Long 1964).
136 Life and History
eighty-two, after advising Augustus well into his reign as emperor (cf.
[Lucian], Macrob. 21).
Of the son of Sandon, we know a good deal historically due to his
association with the young Octavian and the mature Augustus. (A full
and descriptive analysis of the assignment of the biographical ‘Athe-
nodorus’ fragments may be found in Goulet’s Dictionnaire.) In add-
ition to his candidacy as the author of a commentary against the
Categories, Athenodorus Calvus is credited with a ‘history of his
fatherland’; with a work dedicated to Octavia Minor; with a work
 æd
ıB ŒÆØ ÆØ Æ; and with a treatment of at least eight
books on ‘Peripatetics’ (e.g. D.L. 9.42.4, although it is also plausible
that this is Cordylion’s). As the later sources were primarily interested
in Calvus’ association with Augustus, many anecdotes about his
mentorship have come down to us, several of which may carry
hints about his ethical doctrine. There is a famous anecdote that he
instructed Augustus to recite the alphabet whenever he became angry,
before he took any action. Whoever introduced this anecdote may
have intended to reflect, and explain retroactively, certain traditions
about the historical Augustus’ clementia. But the anecdote may also
reflect some specific interest on the part of Athenodorus in the Stoic
doctrine of ‘first motions’ (on which see my chapters on Andronicus
and Boethus).
We know less about the Athenodorus called Cordylion. Strabo
mentions him as a Tarsian Stoic alongside Antipater, Archedemus,
Nestor, and Athenodorus Calvus (Strabo 14.5.14, a list which bears
comparison to the roster of Stoic logicians at DL 7.68).16 During his
tenure as librarian at Pergamon, according to the rhetorician ‘Isidorus
of Pergamon’, Cordylion attempted to excise passages from the
Republic of Zeno: Isidorus presents the consequences in a rather
unflattering light (çøæÆŁ  F ŁÅ æı ŒÆd ŒØ ı 
Æ ,
DL 7.34). By 67 bce, Cordylion had won renown as a Stoic, and had
also developed a reputation for refusing the patronage of kings
and politicians (Plutarch, Cato Minor 10,1–3). But a determined
M. Porcius Cato succeeded in winning him over, and subsequently
Athenodorus went to live with Cato in Rome (Plutarch 16,1; Strabo
14.5.14). He lived until 50–46 bce.

16
The logician ‘Athenodorus’ mentioned by Diogenes Laertius 7.68 is also men-
tioned alongside Antipater and Archedemus.
 Stoic Critique 137
Richard Goulet (in Goulet 2000) exercises caution in assigning
ambiguous fragments to any single Athenodorus, noting the wide
and poorly distinguished field of candidates. But there is some con-
sensus among contemporary specialists, including Goulet, Moraux,
and Hijmans, that the Athenodorus who commented on the Categor-
ies is the son of Sandon—chiefly because of Calvus’ late dating, taken
in association with the date traditionally assigned to the publication
of Andronicus’ edition of Aristotle (see for example Moraux I: 150).
Hijmans presents a list of arguments in favour of this identification,
and is followed in the main by Goulet. Hijmans narrows the candi-
dates to the two Tarsians: the alternative candidates, such as the
‘Athenodorus of Rhodes’ mentioned by Quintilian (2.17.15), are
said to be too obscure, and the earlier student of Zeno and brother
of Aratus, ‘Athenodorus of Soli’, is hardly mentioned in the litera-
ture.17 Second, the Athenodorus cited by Diogenes Laertius (7.68) as
a logician interested in assertion could well be identified with the son
of Sandon (Hijmans 105, citing Zeller’s History, 1883: 607). Third, it
appears probable that Athenodorus Cordylion died before Androni-
cus’ edition of the Categories was published.
The strength of these arguments in favour of the candidacy of
Calvus is perhaps questionable, and I have argued elsewhere18 that
it would be more effective to focus attention upon the ‘Athenodorus’
mentioned by Diogenes Laertius as a pioneering Stoic logician at 7.68,
and to seek reasons internal to the text to identify this Athenodorus
either with Calvus or with any of the other Athenodori in our record.
At 7.68, ‘Athenodorus’ is said to have concerned himself with the
distinction between simple and complex assertion, and to have
included ‘categorical’ propositions (ŒÆ ƪæ ı ØŒ) within the rubric
of ‘simple’ propositions (7.69–70). Why might this Athenodorus be
relevant to the Categories commentaries? He held that a ‘categorical’
assertible (cf. 7.70) such as ‘Dion walks’ is a simple assertible. The
Stoic critic of the Categories mentioned by Dexippus (in Cat. 12,3–11)
contended that Aristotle, in the Categories, fails to account for pro-
positions such as ‘Dion walks’ precisely as simple, categorical asser-
tibles. Dexippus’ response is that this is a complex proposition and
therefore excluded by the scope of the Categories. Of course, the

17
Goulet plainly states that he is unlikely to be the Athenodorus of DL 7.68, and he
is not even mentioned by Hijmans.
18
I argue this in more detail in Griffin 2013b.
138 Life and History
notion of logical ‘simplicity’ is fundamental to the Categories (e.g.
1b25), but in quite a different way: the atomic units of the Categories
are simple terms, whereas the atomic units of Stoic logic are simple
propositions.
That distinction appears to have been played upon for polemical
ends by both critics and defenders of the Categories, as Dexippus
testifies. It may also be suggestive of the motives that led Stoic
logicians such as Athenodorus and Cornutus to focus a series of
commentaries on the problems raised by the Categories. Jonathan
Barnes points out, for example, Alexander’s essay on negation (in
A.Pr. 402,1–405,16), focusing on sentences such as ˚ƺº Æ
 æØÆ E.19 For Alexander, a negation-sign modifies the verb
 æØÆ E alone; but for his opponents (evidently Stoic logicians), a
negation-sign modifies the entire propositional sentence, which
should itself be viewed as ‘simple’. It is generally assumed that
Alexander’s opponents here are ‘old Stoics’, but Barnes makes a
case that ‘they may . . . have been later Stoics —Athenodorus and
Cornutus, say—who determined to defend the old Stoic doctrine
against its resuscitated Peripatetic rival’.20 Comparing the passage of
Dexippus noted above (and discussed in more detail below) adds
some weight, I think, to this association.
For our purposes here, this connection is relevant because it links
the professional interests of ‘Athenodorus’ at DL 7.68 with the criti-
cism broadly ascribed to ‘Athenodorus’ by Porphyry, Simplicius, and
(implicitly) by Dexippus—namely that the Categories omits many
types of ‘categorical’ propositions it ought to include. A specific
example of this failure—alongside many ‘lexical’ samples such as
conjunctions and articles—is its inability to account for simple asser-
tibles among ‘things said without combination’. This, while by no
means a decisive link, is at least a textually grounded association
between our Athenodorus and the logician of DL 7.68. The connec-
tion with the Stoic doctrine about simple assertibles might also offer
some motivation for Stoic distrust of the Categories’ concept of
assertion.
We may also note here that there is limited evidence of Boethus
responding directly to Athenodorus, or articulating the kind of strong

19
Barnes 1999: 41–2; see also J. Barnes, ‘Peripatetic Negations’, OSAP 4 (1986),
201–14.
20
Barnes 1999: 43.
 Stoic Critique 139
and systematic response to his critique that he apparently offered to
Lucius; this might weigh against Athenodorus being a very early
figure already well established by Boethus’ floruit, although this is
by no means decisive (particularly given our uncertainty about the
chronology of the work of Boethus himself). Whether the Atheno-
dorus of the commentators is most plausibly identified with Calvus,
Cordylion, or another candidate, it is clear that the historical record
offers little information of value for augmenting and interpreting the
remarks on the Categories assigned to Athenodorus. The point that
I would like to stress here is that, regardless of historical associations,
the Athenodorus of DL 7.68 is the only figure of that name in the
textual tradition outside the Neoplatonic commentators that we know
to have held views about logic that are relevant to, and compatible
with, the criticisms ascribed to ‘Athenodorus’ by Porphyry and
Simplicius.

Cornutus
In the case of Cornutus, there is only one candidate whose identifi-
cation is broadly accepted. He is L. Annaeus Cornutus (whose prae-
nomen is noted in Charisius, Gramm. 162,9, and nowhere else). For
our purposes in this chapter, this identification will prove especially
valuable as it associates Cornutus with the authorship of the Epi-
drome—a work of etymological exegesis (and not of ‘allegory’, as
A.A. Long has stressed)—which demonstrates Cornutus’ sensitivity
to the Stoic doctrine of linguistic signification, and helps to link the
criticisms assigned to ‘Athenodorus and Cornutus’ with the doctrine
of linguistic meaning ascribed broadly to ‘the Stoics’ by authors such
as Diogenes Laertius and Sextus Empiricus.
Cornutus is chiefly known to the historical tradition as the tutor of
the silver-age poet Persius (Satura 5) and as an associate or instructor
of the emperor Nero (see Cassius Dio 62.29.2 and following). While
Persius’ fifth satire is a well-known idealizing portrait of the philoso-
pher, perhaps the most famous anecdote concerning Cornutus is the
story relayed by Cassius Dio in the passage cited above:
[Nero] was now making preparations to write an epic narrating all the
achievements of the Romans; and even before composing a line of it he
began to consider the proper number of books, consulting among
others Annaeus Cornutus, who at this time was famed for his learning
140 Life and History
( PŒØF Æ  Kd ÆØ fiÆ). This man he came very near to putting
to death and did exile to an island because, while some were urging him
to write four hundred books, Cornutus said that this was too many and
nobody would read them. And when someone objected, ‘Yet Chrysip-
pus, whom you praise and imitate (KÆØ E ŒÆd ÇźE), composed
many more’, the other retorted: ‘But they are useful to the conduct of
men’s lives’ (Iºº’ KŒ E Æ åæ
ØÆ fiH H I Łæø  ø fi K
). So
Cornutus incurred banishment for this.21 (Loeb translation, slightly
modified)
Among other things, this anecdote points to its source’s respect for
Chrysippus, and suggests that Cornutus may have displayed a similar
respect. The exile of Cornutus, following this testimony, occurred
between 63 and 65 ce. He is known to have published a   Å æØŒÆd
å ÆØ and a æe ŁÅ øæ , both of which are mentioned by
Porphyry (in Cat. 86,20f.) as containing arguments against Aristotle’s
Categories. He also composed a — æd Œ H (title in POxy 3649, ed.
1984), a Virgil commentary, and (within the ‘secondary’ subheading
of Stoic dialectic, which overlaps with the discipline that we might call
‘grammar’) a de figuris sententiarum, and an e enuntiatione vel
orthographia. But his major surviving work is the De natura deorum
or  ¯ Øæc H ŒÆ a c  EººÅ ØŒc Ł ºª Æ (Ł øæ Æ codd.)
ÆæÆ  ø , which should likely be categorized as a work of
Stoic ‘physics’.
G.W. Most has discussed, in general terms, the value of Cornutus’
Epidrome for the study of Stoic allegory (1989); more recently, George
Boys-Stones has explored the historical implications of Cornutus’
book (2003) and the function of his text as a work of ethical pedagogy
(2007). It is a useful complete text and a remarkable example of
schoolbook literature in the first century ce.22 A. D. Nock, who
studied this text in detail in 1931,23 suggested that the Epidrome

21
Ææ
Œ ıÇ  b ‰ ŒÆd a H   øÆ ø æ Ø ±
Æ
ıªªæłø K

Ø , ŒÆd  æ ª F ºŁı H غ ø , æd ŒÆd › ØF ÆP H
ı Ł E ÆØ,
K
ŒłÆ , ÆæƺÆg ¼ººı ŒÆd  ÆE ˚æ F  PŒØF Æ  Kd
ÆØ fi Æ. ŒÆd ÆP e Oº ªı b ŒÆd IŒ Ø , K B
 ’ s K ƺ , ‹ Ø Ø H
æÆŒ
ØÆ IØ ø ÆP e غ Æ ªæłÆØ, ºº ÆP a r ÆØ çÅ ŒÆd Å Æ ÆP a
I ƪ 

ŁÆØ, ŒÆ Ø  N  ‘ŒÆd c æ
Ø, n KÆØ E ŒÆd ÇźE, ºf
º ø
ı ŁÅŒ ’ I Œæ Æ  ‹ Ø ‘Iºº’ KŒ E Æ åæ
ØÆ fiH H I Łæø  fi ø K
’. ›
b s ˚æ F  çıªc Kd  fi ø tçº . . . (ed. Boissevain 1895)
22
Most 1989.
23
‘Kornutos’, RE Suppl. 5 (Stuttgart 1931), 995–1005.
 Stoic Critique 141
was in pressing need of a new edition.24 Nock established the links of
the Epidrome with earlier Stoic works, and showed that the text was
largely free from interpolation; work on the manuscript tradition in
the 1970s established the text on a firmer footing.25
For our purposes, however, the most relevant aspects of Cornutus’
Epidrome revolve around his attitude towards etymology and the use
of language. In his paper on ‘Cornutus and Stoic Allegoresis’, Most
explores the Epidrome from the perspective of Stoic exegetical prac-
tice. Cornutus follows Stoic tradition26 and relies chiefly on etymo-
logical analysis in this text;27 like Cleanthes,28 Chrysippus,29 and to a
lesser extent Zeno,30 Cornutus frequently explains mythical names
and epithets in terms of their alleged etymological derivation. Unlike
these earlier Stoics, however, he does not accept a single derivation for
each word; rather, Cornutus marshals alternatives from the tradition
(Most cites 1,1–2,4 and 74,18–21 as examples) and does not char-
acteristically select a single ‘correct’ option or promulgate a specific
science of ‘true’ etymology.
Cornutus’ treatment of the poets, as Gill puts it, presupposes ‘the
Stoic theory that language has “natural” meaning and is not simply a
set of conventional symbols’.31 Long also stresses Cornutus’ depend-
ence on etymology and the doctrine that names have an ‘original
meaning’.32 Boys-Stones has persuasively argued, however, that
there is no recoverable evidence in the Epidrome itself for linguistic

24
Certainly the writer of the Epidrome is an interesting case of an author widely
appreciated by the ancients but largely panned, especially on matters of style, by the
modern commentators. For Aulus Gellius (NA 2.6.1) Cornutus was one of the
grammatici . . . haud sane indocti neque ignobiles; for Macrobius (Sat. 5.19.3) he was
tantus uir, Graecarum etiam doctissimus litterarum.
25
P. Krafft, Die handschriftliche Überlieferung von Cornutus’ Theologia Graeca =
Bibl. der klass. Altertumswiss. N.F., R. 2, Bd. 57 (Heidelberg, 1975).
26
See Long, cited below, and previously F. Buffière, Les Mythes d’Homère et la
pensée grecque (Paris, 1956), 60–5; A. Le Boulluec, ‘L’allégorie chez les Stoïciens’,
Poétique 23 (1975), 306 and following; J. Whitman, Allegory. The Dynamics of an
Ancient and Medieval Technique (Oxford, 1987), 36 and following.
27
Most 1989: 2027; and see Epid. 2,4; 2,9; 65,8 for examples of Cornutus’ own use
of K ıºª Æ and cognates.
28
SVF 1.535, 540–3, and 546–7.
29
SVF 2.1021, 1062–3, and elsewhere.
30
SVF 1.103.
31
C. Gill 2003: 38. See also Long and Sedley 1987: 195, Boys-Stones: 49–59.
32
‘Stoic Readings of Homer’, 53–4 = Stoic Studies, 71–2.
142 Life and History
naturalism33 of the kind that we might derive, for example, from
Chrysippus’ interpretation of etymology (see for example SVF 2.146 =
Origen, Contra Celsum 1.24). In any case, we might reasonably
suppose that for Cornutus, a true linguistic expression (say, ‘Dion is
walking’) signifies a complete ‘sayable’ or ºŒ  , namely the prop-
osition that Dion is walking, which in turn picks out a state of affairs
in the world, namely, the actual Dion in motion. That contrasts
against the Aristotelian semantic theory that interposed concepts, or
‘affections of the soul’, between words and realities (De Int. 16a3).
Stoics took assertibles, or propositions (IØÆ Æ), to be examples
of self-complete º Œ  (cf. Sextus, P.H. 2.104, DL 7.63). One state-
ment of the theory is given by Sextus, Adv. math. 8.11–12:
The Stoics say that three things are linked to one another: what is
signified, what signifies, and what exists. Of these, what signifies is the
sound, e.g. the sound ‘Dion’; what is signified is the very thing which is
suggested by the sound and which we apprehend to subsist with our
thought (the barbarians do not understand although they hear the
sound); and what exists is the external object, such as Dion himself.
Of these things two are bodies, namely the sound and what exists, one is
not a body, namely the thing signified, the º Œ  , which also is true or
false.34 (tr. Mueller 1978)
We understand from the Neoplatonist commentators, including Por-
phyry (in Cat. 59,10–14) and Simplicius (18,28–19,1), that Atheno-
dorus and Cornutus adopted the view that the Categories concerned
‘linguistic expressions as such’ (º Ø ŒÆŁe º Ø). From the stand-
point of such an ‘orthodox’ Stoic theory of dialectic, the locution
º Ø ŒÆŁe º Ø would imply that the Categories concerned the
secondary, purely ‘grammatical’ aspect of dialectic, not the º Œ  or
‘sayables’ that are represented by words. (A good example of the

33
G. Boys-Stones, ‘The Stoics’ Two Types of Allegory’, in Boys-Stones (2003a),
189–216. I am grateful to Boys-Stones for conversations that helped me to temper my
previous assumption of such naturalism in the text.
34
ŒÆd c B b æ Å Å æ
ή
Ø ƒ Ie B  A, æ Æ ç Ø

ıÇıª E IºººØ, 
ÅÆØ   ŒÆd e
ÅÆE  ŒÆd e ıªå  , z
ÅÆE 
b r ÆØ c çø  , x  c ˜ ø ,
ÅÆØ   b ÆP e e æAªÆ e ’ ÆP B
ź  ŒÆd y  E b I غÆÆ  ŁÆ fi B  æÆ fi ÆæıçØ
Æ ı ØÆ  fi Æ, ƒ
b æÆæØ PŒ KÆ"ı
Ø ŒÆ  æ B çø B IŒ , ıªå  b e KŒ e
Π  , u
 æ ÆP e › ˜ ø .  ø b  b r ÆØ
Æ Æ, ŒÆŁ æ c
çø c ŒÆd e ıªå  , £ b I
Æ  , u
 æ e
ÅÆØ   æAªÆ, ŒÆd º Œ  ,
‹ æ IºÅŁ ª ÆØ j ł F (eds Mau and Mutschmann 1914).
 Stoic Critique 143
kinds of subjects falling under the first heading is the work list of
Chrysippus at DL 7.192, where a number of works that we might call
‘grammatical’ are headed  æd a º Ø.) This distinction enabled
the Stoics to critique the Categories as an almost absurdly deficient
treatment of verbal expressions as such, in much the same way as
other sources such as Lucius, Nicostratus, and Plotinus, who inter-
preted its subject matter as ontological, would question its compre-
hensiveness as a treatment of being.
The response of Porphyry to this critique, apparently following
Boethus of Sidon and the Aristotelian Herminus (cf. Porph. 59,16),
is that the true subject of the Categories is ‘simple significant vocal-
izations, insofar as they are significant of realities’ ( æd çø H

ÅÆ ØŒH ±ºH , ŒÆŁe
ÅÆ ØŒÆ N
Ø H æƪ ø ). Porphyry,
as we shall see below, justifies this response in terms of a distinction
between the ‘primary’ and ‘secondary’ impositions of verbal expres-
sions (58,30–59,2). His distinction in fact matches perfectly with the
twofold framework ascribed to ‘the Stoics’ by Diogenes and Sextus,
with his ‘primary imposition’ corresponding to the use of º Œ .
Yet Porphyry ridicules Athenodorus and Cornutus, both Stoics
with professional interest in dialectic, for simply failing to recognize
this division at work in the Categories. He suggests that Athenodorus
and Cornutus are actually ignorant of it:
Q. Has everyone who has written about the Categories been aware of
this distinction? A. Certainly not. Otherwise there would not have been
those . . . who attacked the work and rejected the division of categories
as being insufficiently comprehensive . . . the followers of Athenodorus
and Cornutus, who took the objects of the investigation to be expres-
sions qua expressions.35 (59,5–14, tr. Strange 1992)
Athenodorus and Cornutus, as Porphyry read them, may not have
mentioned the distinction of expressions-as-significant from expres-
sions-in-themselves, or may have ignored it wilfully. But they were

35
{ ¯ .} ἎæÆ s c ØÆçæa Æ Å   ª ø
Æ ƒ  æd H ŒÆ ŪæØH Ø
ªæłÆ ; {.} ˇPÆH· P ªaæ i ƒ b  æd H ª H H Z ø æŪı ø
fiþ  K ÆFŁÆ æƪÆ 
ŁÆØ, ƒ b I º ª IŁ F  ÆP H c ØÆ æ
Ø ‰
ººa ÆæØ E
Æ ŒÆd c  æغÆ ı
Æ j ŒÆd ºØ º  Çı
Æ . { ¯ .}  N
d
y Ø; {.} ˇƒ  æd ŁÅ øæ ŒÆd ˚æ F  ƒ a ÇÅ  Æ  æd H º ø ŒÆŁe
º Ø, x Æ a ŒæØÆ ŒÆd a æØŒa ŒÆd ‹
Æ ØÆF Æ, (ØÆçæÆd ªæ K
Ø º ø ŒÆŁe
º Ø N
) a ØÆF Æ s æçæ  ŒÆd  Æ K
d ŒÆ Ūæ Æ IæF  ŒÆd c
æ
Œ  KººØB çÆ
Ø r ÆØ c ØÆ æ
Ø , ‰ i c 
Å çø B
ÅÆ ØŒB N
ÆP c  æغÆÆ  Å (ed. Bodéüs 2008).
144 Life and History
surely sensitive to the difference, despite Porphyry’s allegation
that Athenodorus and Cornutus were simply ‘ignorant’ of it ( c
ØÆçæa . . . ª ø
Æ . . . PÆH, 59,3–5). Based on our knowledge
of their work, not least of Cornutus’ Epidrome, it is very unlikely that
Athenodorus and Cornutus were simply unaware or naively disinter-
ested in the Stoic division of dialectic into primary and secondary
aspects.36 Rather, Athenodorus and Cornutus must have intention-
ally associated the Categories with the secondary or purely verbal
species of Stoic dialectic, but left that presumption sufficiently impli-
cit or vague for Porphyry to take advantage of it to score his point in
reply. More specifically, Athenodorus (for his particular criticism see
Simplicius in Cat. 62,25)—or both Athenodorus and Cornutus—
must have analysed the first chapter of the Categories as a purely
linguistic or ‘secondary’ treatment of ‘onymies’ (18,22–19,7), and
then proceeded to paint the entire text with the same brush.
The semantic assumptions that underlie Porphyry’s defence
against Athenodorus, so far as they are stated explicitly, are simple:
[Aristotle] adopted the word [ŒÆ Ūæ Æ], and chose to call those
utterances in which significant expressions are applied to things ‘predi-
cations’ (ŒÆ Ūæ ÆØ). Hence whenever a simple significant expression
is employed and said of what it signifies, this is called a ŒÆ Ūæ Æ. For
example, this stone I am pointing at . . . is a thing, and when we say
about it, ‘This is a stone’, the expression ‘stone’ is a ŒÆ Ūæ Æ, for it
signifies that sort of thing, and is uttered about the thing we are pointing
at, the stone. So too in other cases.37 (56,6–13, tr. Strange 1992)
There is nothing especially ‘Aristotelian’ about the semantics here (in
that the underlying theory does not allude, say, to the opening of the
De Interpretatione, featuring ÆŁÆ Æ of the soul mediating between

36
Cornutus’ use of etymology in the Epidrome, and the titles of his other works in
grammar, suggest that he was sensitive to—and, in his pedagogical practice, a sub-
scriber to—this differentiation. Moreover, as I noted above, it appears probable that
the commentators’ Athenodorus (regardless of his historical identity) is also the
logician concerned with simple significant propositions at Diogenes Laertius 7.68; it
is highly unlikely that his argumentation regarding the skopos of the Categories was as
unsophisticated as Porphyry suggests.
37
ºÆg ÆP e a H º ø H
ÅÆ ØŒH ŒÆ a H æƪ ø Iªæ 
Ø
ŒÆ Ūæ Æ æ
E . u
A
Æ ±ºB ºØ
ÅÆ ØŒc, ‹ Æ ŒÆŁ’ y
ÅÆ Å ÆØ
æªÆ  Iªæ ıŁB ŒÆd º åŁB fi , ºª ÆØ ŒÆ Ūæ Æ, x  Z  æªÆ  F F
 ØŒ ı ı º Łı, y ±  ŁÆ j n º , ‹ Æ Yø K’ ÆP F ‹ Ø  º Ł
K
,  º Ł ºØ ŒÆ Ūæ Æ K
Ø·
ÅÆ Ø ªaæ e Ø  æAªÆ ŒÆd Iªæ  ÆØ
ŒÆ a F  ØŒ ı ı æªÆ  º Łı. ŒÆd o ø Kd H ¼ººø (ed. Bodéüs 2008).
 Stoic Critique 145
verbal expressions and real things). As the rest of Porphyry’s account
shows, he relies here more on a theory of the ‘imposition’ of names on
things that is at home in Stoic and Epicurean semantics (though there
is also nothing uniquely Stoic or Epicurean here).38

II. SOURCES FOR TWO ØÆ æ
Ø

IN THE CATEGORIES

We turn now to the direct sources for the commentaries of Atheno-

dorus and Cornutus on the Categories of Aristotle. The main sources
are the commentaries of Porphyry, Dexippus, and Simplicius.
Dexippus 11,1–12,31 likely distils Porphyry’s actual reply to Atheno-
dorus and Cornutus in Ad Gedalium. The early passages in Simplicius
(such as in Cat. 18,22–19,7) may represent Porphyry’s opening dis-
cussion of Athenodorus and Cornutus, and specifically his effort to
situate their criticisms in the context of the skopos of the Categories,
represented also in the commentary By Question and Answer.
Porphyry, Dexippus, and Simplicius ascribe various critical
remarks to Athenodorus, to Cornutus, or to both throughout the
text of their commentaries On the Categories. The most widely
reported group of comments relates to ‘the division’ or ØÆ æ
Ø.
However, as I suggest below, ØÆ æ
Ø is an ambiguous term that was
applied to two distinct lines of criticism, which are preserved in two
(possibly independent) traditions but are amalgamated by Simplicius

38
On the omission of concepts in the shorter commentary, which may be
explained by its introductory nature, see for example Hoffmann 1987: 68–90 and
Ebbesen 1990: 147. In Griffin 2012a, I suggest that the difference could also arise from
different layers in the sources: the ‘conceptless’ account could derive from Androni-
cus, who would have omitted De Interpretatione as a guide to Aristotelian semantics at
any rate (and may instead have deployed Hellenistic resources like the theories of
‘imposition’ rooted in Plato’s Cratylus), whereas the version that includes concepts
might belong to Boethus of Sidon’s critique of Andronicus.
A further possible example of the ‘Stoic’ context of the defence against Atheno-
dorus’ criticisms is found in Dexippus (discussed below), where his example of the
conjunction as a verbal expression omitted by Aristotle is countered by showing that
the conjunction is ‘co-significant of a break in thought’ (IÆæ Ç B ØÆ  Æ). This
‘break in thought’ is a ‘syntactic’ reality signified by a º Œ  , and the terminology is
Chrysippean (cf. SVF 2.184 and see also Sextus PH 2.176,5 and for the grammatical
usage Apollonius Dyscolus De adv. 121,3 and De const. 402,2). But I would not press
this too far.
146 Sources for Two ØÆ æ
Ø in the Categories
in one passage. It is critical to distinguish (I.1) criticisms of the
ØÆ æ
Ø of the ten categories from (I.2) criticisms of the ØÆ æ
Ø
of the Categories as a text into three major sections (namely the
onymies, predicables, and post-praedicamenta). The first criticism is
reliably attributed to Athenodorus and Cornutus, but it is less clear
that they are the source of the second. The word ØÆ æ
Ø appears to
have been used (though possibly later) in both cases.
(1) The first line of criticism is associated with a definition of the

Œ as ‘verbal expressions qua verbal expressions’, and is rebutted
by Porphyry (as reported by Porphyry himself in the QA commen-
tary, and also by Dexippus and Simplicius). According to this criti-
cism, the ‘division’ of ten categories is unable to comprehend the
entirety of significant verbal expression ( c ØÆ æ
Ø , ‰ i c

Å çø B
ÅÆ ØŒB N ÆP c  æغÆÆ  Å, Porphyry
59,14; c ØÆ æ
Ø , ‰ P 
Æ a º Ø  æغÆF
Æ , Simplicius
18,30; cf. Simplicius 62,26).
(2) The second line of criticism asserts the structural incoherence
of the Categories, showing that it combines the disciplines of logic,
physics, ethics, and theology, and perhaps lacks any coherent
Œ;
this criticism is rebutted by another source (likely Iamblichus, but
possibly Porphyry in the lost commentary Ad Gedalium). This criti-
cism focuses on the division of chapter headings ( N a Œ çºÆØÆ
ØÆØæ
ø F غ ı, Simplicius 18,22), i.e. the tripartition of the
major sections into three ( N  ªºÆ æÅ Ø º E , æØ æb e
غ  , Simplicius 19,9). This line of criticism is not ascribed with
absolute certainty to both Athenodorus and Cornutus, as the central
citation of their names might not be associated with the entire
passage.
Simplicius generally distinguishes these two lines of criticism: for
example, (1) is represented alone at 62,26 and 359,1. But some
confusion arises from a single passage where Simplicius amalgamates
both criticisms (18,22–19,9). Here Simplicius uses ØÆ æ
Ø in three
different senses, referring first to the chapter division ( N a Œ çºÆØÆ
ØÆØæ
ø, 18,22), then to the tenfold division of categories, quoting
Porphyry silently (18,27 and 18,30–19,1), and finally to the internal
division of the onymies (ØÆ æ
Ø H O  ø , 19,1), before return-
ing to the chapter division at 19,9 ( N  ªºÆ æÅ Ø º E , æØ æb
e غ  ). As I suggest below, this ambiguity arises from the com-
bination of at least two sources in this passage.
 Stoic Critique 147

Athenodorus and Cornutus on the ‘Division’

of the Ten Categories
Porphyry
The first main text is found in Porphyry’s surviving commentary By
Question and Answer, where he expresses his position (58,3–7) in the
context of a passage justifying the title ‘Categories’. (This text has also
been treated above.)
The subject of this book is the primary imposition of expressions ( æd
B æ Å Ł
ø H º ø ), which is used for communicating
about things ( H æƪ ø ). For it concerns simple significant
words insofar as they signify things (ŒÆŁe
ÅÆ ØŒÆ N
Ø H
æƪ ø ) —not however as they differ from one another in number,
but as differing in genus . . . (12) Since beings are comprehended by ten
generic differentiae, the words that indicate them have also come to be
ten in genus, and are themselves also so classified. Thus predications
(ŒÆ Ūæ ÆØ) are said to be ten in genus, just as beings themselves are
ten in genus.39 (tr. Strange 1992)
Porphyry stresses that Aristotle is concerned, in the Categories, to
describe the primary imposition of expressions, and not the secondary
imposition of expressions. He goes on to reiterate this point and to
contrast the intention of the De Interpretatione:
Q. But if he here divides significant words into ten genera, why is it that
in On Interpretation he divides them into two, namely nouns and verbs
( N Z Æ ŒÆd ÞBÆ)? A. Because here he is discussing the primary
imposition of expressions upon things ( æd B æŪı Å Ł
ø
H º ø B ŒÆ a H æƪ ø ), while in On Interpretation he is
discussing their secondary imposition ( æd B  ı æÆ), which is no

39

Ø  ı  æŁ
Ø F غ ı  æd B æ Å Ł
ø H º ø B
ÆæÆ
Æ ØŒB H æƪ ø · 
Ø ªaæ  æd çø H
ÅÆ ØŒH ±ºH , (5) ŒÆŁe

ÅÆ ØŒÆ N
Ø H æƪ ø , P c H ŒÆ a IæØŁe Iºººø ØÆç æ ø Iººa
H ŒÆ a ª · ¼ ØæÆ b <ªÆæ>
å e ŒÆd a æªÆ Æ ŒÆd ƃ º Ø ŒÆ a IæØŁ .
Iºº’ P a ŒÆ a IæØŁe æŒ Ø ÆØ Ø ºŁ E º Ø· Œ
Å ªaæ ŒÆ a IæØŁe <$ >

ÅÆ Ø H Z ø · Iºº’ K d fiH IæØŁfiH ºº K
Ø £ Z Æ fiH Y Ø j fiH ª Ø, ŒÆd 
I Øæ Æ H Z ø (10) ŒÆd H
ÅÆØ ı
H ÆP a º ø N ŒÆ ª Å oæÅ ÆØ
 æغÆÆ  Å N e ªæç
ŁÆØ. N ŒÆ  ı ª ØŒa ØÆçæa  æغÅçŁ ø
H Z ø ŒÆ ŒÆd ƃ źF
ÆØ ÆF Æ çø Æd ª ª Æ
Ø ŒÆ a ª Å ŒÆd ÆP Æd
 æغÅçŁ E
ÆØ. ŒÆ s ºª ÆØ ŒÆ Ūæ ÆØ fiH ª Ø ź  Ø, u
 æ ŒÆd ÆP a a
Z Æ ŒÆ fiH ª Ø (ed. Bodéüs 2008).
148 Sources for Two ØÆ æ
Ø in the Categories
longer concerned with expressions that signify things qua signifying
them, but rather with expressions that signify types of words, qua being
of such types ( æd H
ÅÆ ØŒH º ø F ı H çø H , ŒÆŁe
Ø N
d Ø ø ). For being a noun or a verb is a type of word, and
whether an expression has its proper use ( e Œıæ Æ ) or is metaphorical
( ÆçæØŒc ) or is in some other way figuratively used (¼ººø
æØŒc ) also belongs to the second sort of inquiry about words, not
to the first.40 (58,30–59,2, tr. Strange 1992)
The theory of two sequential Ł
Ø of language had been formulated
in the Hellenistic schools,41 and within the context of a broader
debate, building around central passages in Plato’s Cratylus (e.g.
421C–6C),42 about whether names were imposed by nature (ç
Ø)
or by Ł
Ø in the sense of convention.43 In the Stoa, at least since
Chrysippus,44 it had been maintained that primary vocalizations
signified æªÆ Æ ‘by nature’; so Origen informs us in a widely
cited passage of the Contra Celsum (ç
Ø Øı ø H æ ø
çø H a æªÆ Æ: Cels. 1.24 = SVF 2.146 = Hülser FDS fr. 643).45

40
{ ¯ .} ºº’ N K Ł N ŒÆ ª Å Ø Eº a
ÅÆ ØŒa çø , (30) H K fiH
— æd æÅ Æ N , N Z Æ ŒÆd ÞBÆ; {.}  O Ø K Ł b  æd B
æŪı Å Ł
ø H º ø B ŒÆ a H æƪ ø Ø E ÆØ e ºª , K
b fiH — æd æÅ Æ  æd B  ı æÆ, m PŒ Ø K
d  æd H
ÅÆ ØŒH º ø
H æƪ ø , ŒÆŁ N
Ø  ø
ÅÆ ØŒÆ , Iººa  æd H
ÅÆ ØŒH º ø F
ı H (35) çø H , ŒÆŁe Ø N
d Ø ø ·  ªaæ B çø B e r ÆØ j
Z Æ j ÞBÆ. ŒÆd e Œıæ Æ b r ÆØ c ºØ j  ÆçæØŒc j ¼ººø [p. 59]
æØŒc B  ı æÆ K
d  æd H çø H æƪÆ Æ ŒÆd P B æ Å (ed.
Bodéüs 2008).
41
In treating the Stoic and Platonic sources of the theory, I draw in particular on
three studies in Frede and Inwood 2005: Allen 2005, Long 2005, and Ebbesen 2005;
Griffin 2012a is a more detailed treatment of my views here. See also Sedley 1998,
Frede 1989: 2088–9 and 1987: 301–62, and Boys-Stones 2001.
42
On the relevance of the Cratylus see Long 2005 and Sedley 1998. The first
segment of this part of the Cratylus seems to discuss Porphyry’s ‘first imposition’,
but the second segment, where Socrates begins to discuss the theory of language,
corresponds to Porphyry’s ‘second imposition’. For the first imposition, see for
example H æ ø O  ø , 424B; æH  ÆP fiH fiH åæÆ Ø ŒÆd fi B çø fi B PŒ

Ø P
Æ, 423E.
43
For this formulation of the question, see Sextus, Adv. Math. 1.143–4, cf. 37; Adv.
Math. 11.241–2; P.H. 3.267–8; Aulus Gellius 10.4; Simplicius in Cat. 40,6 and 187,7;
and Origen, Exh. Mart. 46.
44
Chrysippus had already discussed the Ł
Ø H O  ø : see Eusebius, PE
6.8.1–10 = SVF 2.914; the phrase, as Allen 2005:18 argues from 6.8,11–24, seems to be
Chrysippus’ own.
45
Platonists seem to have adopted this view, presumably (as in late antiquity) as an
exegesis of the Cratylus. See Alcinous Intr. ch. 6; Proclus in Crat. 16.18 and 18.1–4
Pasquali; Stephanus in Int. 9,19–22 and 10,7; and see Cratylus 390A. The Epicureans
 Stoic Critique 149
For the Stoics, as for Porphyry, the primary Ł
Ø ‘by nature’ tran-
spired in human prehistory.46
Boethus of Sidon and Herminus are later cited as originators of this
assessment, or at any rate as ‘agreeing’ with the line of interpretation
Porphyry has just laid out (ŒÆd BÅŁ K E N a ˚Æ Åªæ Æ
YæÅŒ ÆF Æ ŒÆd  EæE  æÆåø, 59,17–18). But Porphyry reports
that several commentators took a different approach (59,5–14),
exactly by failing to distinguish the primary from the secondary
imposition of expressions upon things:
The followers of Athenodorus and Cornutus . . . took the objects of the
investigation to be expressions qua expressions, that is, expressions as
used properly and figuratively ( a ŒæØÆ ŒÆd a æØŒa) and so forth,
for these are differentiae of expressions qua expressions. Fixing upon
these, they raised the question of what category they belonged to, and
finding none, they complained that the division (ØÆ æ
Ø) was incom-
plete, since it fails to include every sort of significant expression.
(tr. Strange 1992)47
For a detailed treatment of the arguments offered by Athenodorus
and Cornutus to support this criticism, we must turn to Simplicius
(who names them explicitly) and Dexippus (who does not).

Simplicius
Simplicius uses a similar manner of introducing Athenodorus
and Cornutus, situating their arguments in the context of a debate
regarding the skopos of the text. Porphyry certainly lies behind
Simplicius’ treatment, but it is not entirely clear whether Simplicius
is primarily reliant on the Ad Gedalium, or had direct access (as
Strange suggests) to the commentary By Question and Answer, or
has access to Porphyry via Iamblichus. In the next section, I will

also seem to have presented a narrative in which the primordial imposition of

language was ‘natural’, although the sense was very different than that applied by
the Stoics; perhaps what evolved naturally, like the faculty of speech, did so without
‘purpose’, until we found a use for it. See Lucretius 4.824, and Epic. Ep. Hdt. 75.
46
With references above, see Frede (1989:2088–9) on Stoic ‘prehistory’.
47
ˇƒ  æd ŁÅ øæ ŒÆd ˚æ F  ƒ a ÇÅ  Æ  æd H (10) º ø ŒÆŁe
º Ø, x Æ a ŒæØÆ ŒÆd a æØŒa ŒÆd ‹
Æ ØÆF Æ, (ØÆçæÆd ªæ K
Ø º ø ŒÆŁe
º Ø N
) a ØÆF Æ s æçæ  ŒÆd  Æ K
d ŒÆ Ūæ Æ IæF  ŒÆd c
æ
Œ  KººØB çÆ
Ø r ÆØ c ØÆ æ
Ø , ‰ i c 
Å çø B
ÅÆ ØŒB N
ÆP c  æغÆÆ  Å (ed. Bodéüs 2008).
150 Sources for Two ØÆ æ
Ø in the Categories
discuss these questions in the broader context of the Simplician
passage. Here we quote his citation of Porphyry (Simplicius in Cat.
18,26–19,1) directly:
What is more, some contradict Aristotle and reject his division
(IŁ F  c ØÆ æ
Ø ): of these, some claim that it is uselessly
redundant, others that it omits many things (ƒ b ‰ ººa Ææ E
Æ ),
like Athenodorus and Cornutus, who believe that the
Πconcerns
expressions insofar as they are expressions. They bring forward many
expressions as examples, some literal (Œıæ Æ), some figurative
( æØŒ), and thereby think they refute the division, since it has not
included all possible expressions.48 (tr. Chase 2003)
The survey of Athenodorus’ and Cornutus’ views paraphrases Por-
phyry, as we find by comparing the corresponding passage from his
commentary By Question and Answer:49

Porphyry 59,10–14 Simplicius 18,28–19,1

ˇƒ  æd ŁÅ øæ ŒÆd ˚æ F  ƒ a ˚æ F  ŒÆd ŁÅ øæ, ¥ Ø   æd

ÇÅ  Æ  æd H º ø ŒÆŁe º Ø, x Æ
a ŒæØÆ ŒÆd a æØŒa ŒÆd ‹
Æ
ØÆF Æ . . . ‰ i c 
Å çø B

ÅÆ ØŒB N ÆP c
 æغÆÆ  Å . . .
º ø N Ø e
Œe r ÆØ ŒÆŁe
º Ø N
, ººa º Ø æºº  a
b Œıæ Æ, a b æØŒ, Kºªå Ø
Y ÆØ c ØÆ æ
Ø . . . . ‰ P 
Æ a
º Ø  æغÆF
Æ . . .

Previously at 17,3–7, Simplicius claimed to recite the ‘very words’ of

Porphyry, but gave a version slightly different in wording from our
text of the commentary by Question and Answer. Similarly here at
18,28–19,1, Simplicius delivers a nearly verbatim citation of Porphyry
by Question and Answer. This may suggest either that he is citing
our commentary from memory, or via an intermediary (such as

48
ŒÆd c ŒÆd I غªı
Ø ÆP fiH Ø  IŁ F  c ØÆ æ
Ø , ƒ b ‰
º  Çı
Æ  Å , ƒ b ‰ ººa Ææ E
Æ u
 æ ˚æ F  ŒÆd ŁÅ øæ,
¥ Ø   æd º ø N Ø e
Œe r ÆØ ŒÆŁe º Ø N
, ººa º Ø
溺  a b Œıæ Æ, a b æØŒ, Kºªå Ø Y ÆØ c ØÆ æ
Ø , ‰ P

Æ a º Ø  æغÆF
Æ · (ed. Kalbfleisch 1907).
49
Notably, this comparison also suggests that the locution ƒ  æd X was not read
as historically or biographically significant by the commentators, for Simplicius (or
Iamblichus before him) has simply reduced Porphyry’s ˇƒ  æd ŁÅ øæ ŒÆd
˚æ F  to ˚æ F  ŒÆd ŁÅ øæ.
 Stoic Critique 151
Iamblichus), or that he is citing the Ad Gedalium, which might put
the same argument in slightly different words.
Simplicius expands substantially on this criticism of the ‘division’
later in his analysis, at 62,25 and following:
Many others disputed [the division into ten], denouncing immediately
the division into such a multitude; as did Athenodorus in his book
which, although it was entitled Against Aristotle’s Categories, only
investigated the division into such a multitude. Both Cornutus, more-
over, in the work he entitled Against Athenodorus and Aristotle, and
Lucius and Nicostratus, and their followers spoke out against the
division, as they did against practically everything else.50
In the next lines, Simplicius proposes to ‘take up the opposing
arguments in definite terms, making a threefold distinction of them:
for some reproach the division with being excessive, others criticize it
as being deficient; and a third group is constituted by those who
consider that some genera have been introduced in the place of
others’ (62,30–1). He proceeds to sketch these three groups of criti-
cism as follows:
1. 63,4–64,12. Those who accuse Aristotle of excess. (i) Some
[unnamed] argue that Aristotle should have combined Ø E
and 
å Ø under a single heading, ŒØ E
ŁÆØ. (ii) Others [Xe-
nocrates and Andronicus] include the ten categories under two:
ŒÆŁ Æ  and æe ; others still under substance and accident.
2. 64,13–66,15. Those who accuse Aristotle of deficiency. (i) The
followers of Nicostratus assert that Aristotle, if he distinguished
Ø E from 
å Ø , should have distinguished å
ŁÆØ from
å Ø as well. (ii) The followers of Lucius ask why he omitted
conjunctions and articles, which are significant º Ø [Simpli-
cius answers that they ‘co-signify’,
ı

ÅÆ ı
Ø ] and simi-
larly negations, privations, the moods of verbs, and the monad
and the point.

50
@ººØ b ººd æe ÆP c Iç
 Å
Æ , ÆP Ł ŒÆ ŪæF  B N

F  ºBŁ ØÆØæ
ø, u
 æ ŁÅ øæ K fiH —æe a (25) æØ
 ºı
b ˚Æ Åªæ Æ Kت ªæÆ fi ø غ fi ø,  Å b c N 
F  ºBŁ ØÆ æ
Ø
K Ç Ø. ŒÆd ˚æ F  b K x  —æe ŁÅ øæ ŒÆd æØ
 ºÅ KªæÆł ŒÆd
ƒ  æd e ¸ŒØ b ŒÆd e ˝ØŒ
æÆ  , u
 æ æe a ¼ººÆ  Æ
å  , o ø
ŒÆd æe c ØÆ æ
Ø I ØæŒÆ
Ø (ed. Kalbfleisch 1907).
152 Sources for Two ØÆ æ
Ø in the Categories
3. 66,16–31. Those who accuse Aristotle of substitution
(I ƺºÆª). They contend that Œ Å
Ø should be taken up in
place of Ø E and 
å Ø .
Immediately after this third group is described, Simplicius introduces
‘the Stoics’ (ƒ  ª  øØŒ ):
4. 66,32–67,9. ‘The Stoics’ believe that the number of primary
genera should be ‘reduced’ ( N Kº  Æ
ı
ºº Ø ) to four,
some of which they take over in an altered form (źºÆª Æ).
Simplicius proceeds (67,9–69,1) to describe how one should go
about responding to these critics, and in particular how the tenfold
division is validated by Iamblichus and by ‘Archytas’.
Before examining this grouping of critics as a whole, it is worth
noting that Simplicius’ discussion of ‘the Stoics’ does not seem to
continue his third class.51 If anything, the desire of ‘the Stoics’ to
‘reduce’ the number of categories would seem to belong naturally
to Simplicius’ first class of critics, but they are cited out of order and
do not really belong to the classification at all. Thus it seems likely
that Simplicius has finished with his source for the threefold division
of critics, and is now introducing a different source. There is also no
reason whatsoever to assume that ƒ  øØŒd are specifically ƒ  æd
ŁÅ øæ ŒÆd ˚æ F  ; rather, the position associated with ‘the
Stoics’ here is a position generically associated with the Stoic school.
This new source for ‘the Stoics’ is quite possibly not a Stoic at all; it
could well be a third-hand report of Stoic orthodoxy, coupled with a
third-hand observation about how the Stoic school also propounded
a doctrine about ‘the number of primary genera’ ( e H æ ø
ª H IæØŁe , 66,32), which they divided into four. This passage
does not demonstrate that any particular Stoic actually arrived at this
view in the course of criticizing Aristotle’s Categories, or used it as an
instrument to criticize Aristotle with. Thus there is no reason to
deduce from this passage that Athenodorus’ criticism of ‘the division
into such a multitude’ (62,25) actually referred to an excessive number
of categories, viz. ten instead of four;52 all the other evidence suggests

51
It is true that they alter (źºÆª Æ) Aristotle’s categories, but the main
‘action’ recommended by ƒ  ØŒ is N Kº  Æ
ı
ºº Ø , not just substitution.
52
Indeed, there is no evidence that the Stoics’ fourfold division of primary genera
came into the commentaries of Athenodorus and Cornutus at all; if they had, one
would hope that Simplicius and his sources would tell us something about that.
 Stoic Critique 153
that Athenodorus criticized the ten categories as somehow deficient
in number (namely, because they failed to encompass many parts of
speech).
But this does not help us to identify any one of Simplicius’ three
main classes of critics with Athenodorus or with Cornutus. As their
names were cited explicitly at the outset of the tripartition, we should
expect to find their views represented by one of the three groups.
Curiously, all three of Simplicius’ groups launch their argument from
a shared and rather specific Iæ Æ: why did Aristotle not amalgam-
ate Ø E and 
å Ø into a single category, ‘change’? The first group
introduces this complaint (he should have amalgamated them); the
second group, initially identified as the ‘Nicostratans’, elaborates
upon it (he is inconsistent in failing to amalgamate Ø E with

å Ø because he does not distinguish å Ø and å
ŁÆØ); and the
third group essentially repeats the question of the first with a trivial
change of language (he should amalgamate Ø E and 
å Ø under
Œ Å
Ø). A number of seemingly independent commentators are
classed under this tripartition for pedagogical convenience, including
Xenocrates and Andronicus (who seem to have an entirely unrelated
reason for reducing the categories to two, substance and accident)
and Lucius (who points out various º Ø that Aristotle has omitted,
such as conjunctions).
I will return below to the question why all three lines of criticism
begin from the distinction of Ø E and 
å Ø . First, however, we
should try to identify where within the tripartition of positions
Athenodorus or Cornutus can be identified. If no names were pro-
vided, we would almost certainly assign to Athenodorus and Cornu-
tus the view of the second group, expounded from 64,18–65,13 and
explicitly assigned to ‘the Lucians’. (1) First, this position holds
that the Categories is concerned with significant º Ø; we are told
by Porphyry and by Simplicius that this was the position of Atheno-
dorus and Cornutus, whereas Lucius elsewhere seems to maintain the
incompatible view that the Categories is concerned with the genera of
being, an apparent inconsistency for which he has even been

Instead, I think that Athenodorus introduced an argument, as Porphyry and Simpli-

cius seem to make clear, that the number of categories was inadequate to encompass
the entirety of verbal expression.
154 Sources for Two ØÆ æ
Ø in the Categories
reproached.53 (2) Secondly, the examples provided for the omitted
º Ø are just the examples supposedly offered by Athenodorus and
Cornutus, such as conjunctions, articles, and so on. (3) Thirdly, the
unusual verb
ı

ÅÆ ı
Ø used here and first introduced by
Plotinus in a related passage (Enn. 6.1.5,14) is also found in the
answering passage of Dexippus (32,18–29); but it is introduced by
Dexippus at 11,11, in a section headed ‘How not to be led astray by the
difficulties raised by the Stoics’54 and associated with the comments
ascribed to Athenodorus and Cornutus by Porphyry and indeed by
Simplicius himself. (4) Fourthly, the actual structure of Simplicius’
discussion is echoed in Dexippus 32,18–29, where Dexippus’ response
is that the use of conjunctions is ‘not a primary, but a secondary, use of
language’; so too Simplicius 64,20 (h æŪı Å ÆP H K
Ø 

ÅÆ
Æ, Iººa
ı

ÅÆ ı
Ø ). To this language compare Porphyry
58,30 and following ( æd B æŪı Å Ł
ø H º ø B
ŒÆ a H æƪ ø Ø E ÆØ e ºª ): while it is unclear to whom
Porphyry is replying in this portion of the QA commentary, it is
‘the followers of Athenodorus and Cornutus’ upon whom he next
comments by name (59,10 and following), including them among
those who can be refuted by this distinction. Porphyry’s defence is
specifically levelled against their description of the
Œ as º Ø
ŒÆŁe º Ø.55
Despite these considerations that might lead us to credit Atheno-
dorus with this complaint, Simplicius ascribes it to ƒ  æd e
¸ŒØ . Yet there are additional curiosities associated with the
text. (5) Fifthly, Lucius is unusually named apart from Nicostratus
(see ch. 4), and most oddly, he is here named immediately after
Nicostratus with a differing view. Simplicius never elsewhere sets
‘Lucius’ and ‘Nicostratus’ against one another in this way, nor in
this order. (6) Sixthly, immediately after the ascription to Lucius, but

53
Moraux suggests that Lucius is in fact being inconsistent (1984: 532 f.) See also
Chase 2003, n. ad loc.
54
The chapter headings, of course, are almost certainly not by Dexippus. But
J. Dillon, in his translation, comments that they are reasonably accurate, and in this
case the heading is a good guide to the associations assigned by the later tradition to
this section.
55
Porphyry observes that this defence also applies to ‘those who take the investi-
gation to be primarily about the genera of being’ (59,5)—but if that is meant to apply
to the ‘Lucians’, it is all the stranger for Simplicius to assert, in a corresponding
passage, that Lucius believes the Categories to be about lexeis and not about genē tōn
ontōn at all.
 Stoic Critique 155
before the Porphyrian answer that is elsewhere associated with the
Stoics, the text of Simplicius has a lacuna,56 which likely contained
just the beginning of the argument distinguishing the ‘primary’
meaning of some º Ø from the ‘co-signification’ of others, as
preserved in Dexippus. The lacuna certainly need not cast doubt on
the soundness of the surrounding text. It does show, however, that
there may be problems in the manuscript tradition associated with
the ascription to Lucius, and it is a significant coincidence that the
lacuna intervenes right between that ascription and the argument
elsewhere associated with Athenodorus and Cornutus. Moreover, in
the answering passage of Porphyry (86,20–32), where a similar tri-
partition of views is expressed, only Athenodorus and Cornutus are
mentioned explicitly.
Coupling (5) the peculiarity of Lucius’ appearance ‘opposite’
Nicostratus and (6) the uncertainty of the surrounding manuscript
tradition with (1) the incompatibility of the position ascribed to
Lucius at 64,18–65,13 with his views on the skopos as recorded
elsewhere, and considering the positive identification of the cited
view with that elsewhere accorded to Athenodorus and Cornutus,
perhaps we could suggest that the name of either Athenodorus or
Cornutus, or both, was originally present alongside, or in place of, or
just after, the name of Lucius at 64,18. It is certainly possible that
several of the views expressed here—especially those on the monad
and the point (which open a new argument about the distinction
of intelligible from sensible reality from 65,13 onwards)—should
be credited to Lucius, although Dexippus deals with them under the
same heading. But those preceding questions which focus on the
inadequacy of the ten categories to comprehend all lexeis, and
which are rebutted by the Porphyrian tradition with reference to
the distinction of ‘primary’ from ‘secondary’ impositions of verbal
expressions, should probably be ascribed to Athenodorus or to Athe-
nodorus and Cornutus.

Dexippus
Dexippus in Cat. 11,7–12 and the related passage 32,18–29 are
relevant in exploring the attribution of Simplicius 64,18–65,13.

56
As Chase 2003 observes, this lacuna was already present in the edition used by
Moerbeke.
156 Sources for Two ØÆ æ
Ø in the Categories
Dexippus tackles the (evidently Stoic) argument that the Categories
should include conjunctions among other ‘grammatical’ elements by
remarking that the Categories considers only what is significant—and
conjunctions are ‘co-significant’:
If we take an element of speech which is non-significant in itself, such as
blityri, or if something is significant by reference to something else, as in
the case of so-called ‘pronouns’ . . . or if a term is co-significant
(
ı

ÅÆ ØŒe ) with something else, as is the case with articles and
conjunctions ( a ¼æŁæÆ ŒÆd ƒ
 
Ø), in no way would it be proper
to include these among the predicates (K ÆE ŒÆ Ūæ ÆØ).57 (Dex-
ippus in Cat. 11,7–12, tr. Dillon 1990)
This term ‘co-significant’ (
ı

ÅÆ ØŒe ) is unusual. We first
encounter it in Plotinus’ treatment of the categories (Enn. 6.1.5,14),
where voice is divided into the impact on air and the movement, one
of which signifies while the other ‘co-signifies’. It recurs later in
Dexippus at 32,18–29, the passage mirroring the text of Simplicius
ascribed to Lucius above (which, I have argued, belongs rather to
Athenodorus). Here we are given much more detail about the reason
why conjunctions are such ‘co-signifiers’—they specifically co-signify
‘breaks in thought’ (
ı

ÅÆ ı
Ø e IÆæ Ç B ØÆ  Æ, 32,22),
but on their own they mean nothing. The view that conjunctions ‘co-
signify breaks in thought’ (
ı

ÅÆ ı
Ø e IÆæ Ç B ØÆ  Æ)
is Stoic; IÆæ Ç B ØÆ  Æ is a º Œ  and the terminology is
Chrysippean (cf. SVF 2.184 and see also Sextus PH 2.176,5 and for the
grammatical usage Apollonius Dyscolus De adv. 121,3 and De const.
402,2). The argument is that conjunctions co-signify certain º Œ .
Thus I believe there is some evidence here, once again, that a primary
defence against Athenodorus was made within a distinctively Stoic
theoretical framework—and perhaps Cornutus, in his work Against
Athenodorus, is a promising candidate for its source.
Dexippus’ entire treatment at in Cat. 11,1–16,13 takes the form of a
thorough response to the contention that the Categories concerns
º Ø yet omits certain types of expression, such as conjunctions.

57
PŒF Y Ø ¼
Å K
Ø æØ ŒÆŁ’ Æ e x  e º ıæØ j ŒÆ a I Æçæa Kç’
 æ
ÅÆ ØŒe ‰ ƃ o ø ŒÆº ÆØ KÆ ÆçæÆ , Æ¥ Ø  ŒÆ a c Kd a
IæØ
Æ æØÆ I Æçæa źF
Ø ‰  KŒ E  çø c Kd e d I ƪ Å, j Y Ø

ı

ÅÆ ØŒe YÅ  Ł’  æı (10) ‰ a ¼æŁæÆ ŒÆd ƒ
 
Ø, PÆH a ØÆF Æ
K ÆE ŒÆ Ūæ ÆØ Ł Å ¼ Ø ØŒÆ ø (ed. Busse 1888).
 Stoic Critique 157
This treatment follows Porphyry in essentials.58 Coupled with the
introduction of Athenodorus and Cornutus in Porphyry’s commen-
tary By Question and Answer and with Simplicius’ testimony, this text
provides the clearest outline of the actual arguments mustered by
Athenodorus, Cornutus, or both against the tenfold division of cat-
egories. The chapter heading assigned to 11,1–16,14 is ‘how to iden-
tify the categories and not be led astray by the difficulties raised by the
Stoics’ (—H åæc a ŒÆ Ūæ Æ æ
Œ Ø ŒÆd c ª
ŁÆØ ÆE
H  øœŒH Iæ ÆØ; 1,9). Evidently the author of this heading
meant it to apply to the entirety of 11,1–16,14, although the section
naturally falls into two halves: Dexippus’ long argument from
11,4–12,30, and the independent exchange by ‘question and answer’
from 13,1–16,14.
The first question raised by Dexippus’ interlocutor Seleucus
(11,1–4) is how we can distinguish those expressions (º Ø) that
do fall under the ‘categories’ of Aristotle, and so exclude other º Ø
that are not properly predicative. Dexippus approaches the problem
by offering a definition of ŒÆ Ūæ Æ, then demonstrating that certain
problematic cases of º Ø—those raised by the critics as embarrass-
ments for Aristotle’s classification—are excluded by the definition.
Here is a rough summary of Dexippus’ answers to the problems
raised by Seleucus, which are worth including in full:

Part I: That certain problematic º Ø are excluded

by a proper definition of predication.
1. (11,4–11) We must first identify (a) the genus of ‘the significant’
( e
ÅÆ ØŒe A ª ) and distinguish this from (b) the
non-significant (¼
Å ), such as ‘blityri’, and from (c) the co-
significant (
ı

ÅÆ ØŒe ), such as articles and conjunctions.
2. (11,12–17) A true ŒÆ Ūæ Æ enables us ‘to show the realities to
one another’ ( a æªÆ Æ źF IºººØ) and is thus distin-
guished (i) by its relation to a reality (æe e æAªÆ
å
ø),
(ii) by its differentiation from realities ( B ŒÆ a a æªÆ Æ
ØÆçæA), and (iii) most properly of all, by its capacity to
receive form ( NØ E
ŁÆØ). This definition excludes the fol-
lowing kinds of ºØ.

58
See Hadot 1990.
158 Sources for Two ØÆ æ
Ø in the Categories
2.1 (11,18–20) If a ºØ draws its defining characteristic from
its relation to its own parts (
åÅÆ Ø
ŁB fi . . . E NŒ Ø
æ
Ø e Kغº Æ åÆæÆŒ BæÆ Içæ Ç Ø ) it lacks the
primary signification of logos and it is not a ŒÆ Ūæ Æ.
Examples of such excluded º  Øare ØŁ , YŒŁ ,
ŒººØ
Æ, OæŁ Æ Æ,
ç Æ Æ, and ØÅ ØŒ Æ Æ.
2.2 (11,20–5) And if a ºØ draws its meaning from the rela-
tionship of the objects of thought to each other (ή a c
H Å ø æe ¼ººÅºÆ KغŒ ) it is again not a
ŒÆ Ūæ Æ. Examples of such excluded lexeis are the conse-
quence of the hypothetical proposition ‘If it is day . . . ’ and
the alternative contradiction of the disjunctive proposition,
‘Either it is day . . . ’
2.3 (11,25–30) Non-verbal movement of thought such as
groans and roars (Kd H
ƪH ŒÆd æıåÅ ø ),
and indiscernible sounds (Iç Æ
 çø c), and names
that represent no reality (Z Æ Åb æAªÆ ź
Ø ),
are not ŒÆ Ūæ ÆØ.
2.4 (11,30–12,3) There is no higher predicate, neither real nor
linguistic (k j º ª  ), that may be predicated syn-
onymously of the highest genera of reality nor of the most
generic lexeis. Therefore, ‘Being’ ( e k ) cannot be predi-
cated as common to the highest genera (it is homonymous:
13,5), nor can ‘Motion’ or ‘Change’ (Œ Å
Ø) be predicated
as common to Ø E and 
å Ø .
2.5 (12,3–11) The defining characteristic of ŒÆ Ūæ ÆØ is sim-
plicity ( e ±ºF ) and non-compositeness (I
 Ł  ).
Therefore we must exclude composite expressions, quasi-
composite expressions, invented words, and other expres-
sions that belong properly to the study of verbal expression,
which is secondary and not primary (¼ººÅ Ł øæ Æ B
 æd c ºØ ŒÆd  ı æÆ). Examples of such excluded
lexeis are ˜ ø  æØÆ E and ı
æø (sic: perhaps
ı
æ Ø with Dillon?).
2.6 (12,11–19) A ŒÆ Ūæ Æ must identify the primary signifi-
cation of utterances ( c æŪı Å
ÅÆ
Æ , cf. Por-
phyry 58,30; Simplicius 64,20) and the primary essential
commonalities of beings ( a æŪı Æ P
Ø Ø
ŒØ  Å Æ), but not the secondary common concepts of
them which are accidentally manifested. These include
derivative forms and secondary concepts.
 Stoic Critique 159
3. (12,19–26) This is the approach of ‘serious students of ºª
and of P
Æ’ (
ıÆÇ ÆØ e H F ºªı ŒÆd B
P
Æ çغŁ Æ ø ). Based on these guidelines, we must
omit the following from consideration as categories, and as
not conforming to the skopos of the Categories (
ŒfiH H
ŒÆ ŪæØH , 12,26):
3.1 (12,20–1) Figurative and metaphorical expressions ( e
æØŒe j  ÆçæØŒe Z Æ) such as e ‘I åÆ Ø
’
or ‘Æ Æ   ”Å’.
3.2 (12,22–3) Modal expressions (ƒ æØ) such as e K
I ªŒÅ and K å
ŁÆØ, and æå Ø .
3.3 (12,23–4) Quantifiers (æ
ØæØ
d) such as e A and
P d, and e d, and På Ø.
4. (12,26–31) The Categories are concerned with ºª that
encompasses truth and falsehood and so with ‘such portions
of significant speech as, being taken together to form a simple
predicative statement, give proper value to the terms out of
which they are constructed’. (Notably this formulation differs
slightly from that of Porphyry.)

Part II: Various Iæ ÆØ concerning predications

(to what category does X belong?)
5. (13,3–6) Being belongs to no category, but is said homony-
mously in ten ways.
6. (13,7–10) ‘One’ belongs to no category, but is said homony-
mously in ten ways.
7. (13,11–15) Plurality belongs to no category, but is said homo-
nymously in ten ways.
8. (13,16–23) Not-being belongs to no category. (i) Proper not-
being is non-existent. (ii) Relative not-being is said homony-
mously in ten ways.
9. (13,24–7) ‘Is’ belongs to no category, but is said homony-
mously in ten ways.
10. (13,28–14,2) ‘Same’ and ‘Other’ belong to no category, but are
said homonymously in ten ways.
11. (14,3–7) Each of the categories is ‘something’ ( Ø) homony-
mously in ten ways.
160 Sources for Two ØÆ æ
Ø in the Categories
12. (14,8–19) ‘Definition’ (‹æ) reveals the essence ( e K
Ø) of
a thing by indicating its distinctive characteristic (NØ Å), and
this is said homonymously in ten ways.
13. (14,20–31) ‘Genus’ and ‘proprium’ (N ı) are said homony-
mously in ten ways, as are differentia and species.
13.1 (14,32–15,13) Defence against certain counter-arguments
from the text of Aristotle. (i) Q. How can genus be
homonymous in ten ways if Aristotle offers a single
definition of it (Top. 1.4, 102a31)? A. Because the defin-
ition itself may be ambiguous. (ii) Q. How can ‘same’ be
homonymous in ten ways if Aristotle defines it as said in
only three ways, namely genus or species, or number
(Top. 7.1, 152b31 and following)? A. Because each of
these three is homonymous in ten ways.
14. (15,14–22) ‘Two ways’, ‘three ways’, ‘in two’, ‘in three’, ‘in four’
( e ØåH ŒÆd æØåH ŒÆd e ØåŁ, æØåŁ, æÆåŁa) are
secondary and not primary significant utterances and therefore
need not belong to the categories.
15. (15,23–31) ‘Whole’ and ‘part’ ( e ‹º ŒÆd e æ) also
belong to the secondary use of language and therefore need
not belong to the categories. (If they do, however, they are
relatives.)
16. (15,32–16,2) The phrases ‘universal’ and ‘once for all’, and
‘wholly’ ( e ŒÆŁºı ŒÆd e ŒÆŁÆ ŒÆd e ¼æÅ ) are
dependent upon the categories, in which they participate.
17. (16,3–11) ‘Same’, ‘other’, ‘proprium’, ‘incompatible’, ‘com-
mon’, ‘indefinite’, ‘ambiguity’, ‘synonymy’ ( e ÆP e ŒÆd e
 æ ŒÆd e YØ ŒÆd e Iºº æØ ŒÆd e ŒØ e ŒÆd c
IØÆçæ Æ  Içغ Æ ŒÆd c
ı ø ı Æ ) are said
homonymously of all the categories, and also, as they describe
names and concern the secondary use of language, properly
‘fall outside’ the categories.
18. (16,11–13) Having answered these questions, we may now
turn to the study of the categories ‘at the highest level’ (æe
c º ø  Å KØ
Å ).
From this series of thrusts and counter-thrusts, we can derive
several important observations. The first useful point is the general
 Stoic Critique 161
context of the argument, which establishes that at least some of the
criticisms here originated with Porphyry’s Athenodorus, Cornutus, or
both. The general line of attack throughout Part I is that a number of
‘significant’ º Ø are omitted from Aristotle’s ‘categories’, including,
for instance, articles and conjunctions (11,4–11) and figurative
expressions (12,20–1). We know from Porphyry that this was pre-
cisely the line of attack adopted by both Athenodorus and Cornutus,
and perhaps especially associated with Athenodorus (Simplicius
62,25). The author of the chapter headings in Dexippus clearly
made the same connection. The defence against this argument is
also that adopted by Porphyry against Athenodorus and Cornutus:
it is critical to distinguish the primary from the secondary imposition
of expressions on things (notably itself a Stoic doctrine), and the truly
significant from the merely ‘co-significant’ or
ı

ÅÆ ØŒ . If we do
so, the problematic º Ø will be excluded from consideration as
predications. Having thus defended against many examples of ver-
bally significant º Ø, Dexippus next tackles a host of philosophical
terms that appear to be omitted from the ‘categories’ (13,3–16,13),
mostly by pointing out that they are said homonymously in ten ways
and belong to no category, occasionally with the aid of a reminder
about the secondary and purely linguistic imposition of expressions.
This is a broader criticism to defend against.
The second useful point is the actual specifics of the examples given
by Athenodorus and Cornutus to support their criticisms of the text.
Looking at the examples above, we know that either or both of them
criticized the Categories along the following lines (for the scope of
Part I above):
 Expressions relating to movement from place, such as ØŁ and
YŒŁ , are wrongly omitted. (Compare Cornutus at Simplicius
359,1 and following: should ØŁ at Dexippus 11,18–20 be
amended to the rarer ˜Øø Ł of Simplicius 359,4?)
 Superlative expressions, such as ŒººØ
Æ, OæŁ Æ Æ,
ç Æ Æ,
and ØÅ ØŒ Æ Æ, are wrongly omitted.
 Stoic hypothetical propositions are wrongly omitted, such as ‘If
it is day . . . ’ (compare DL 7.69 and Sext. Emp. Adv. Math.
8.223).
 Stoic disjunctive propositions are wrongly omitted, such as
‘Either it is day . . . ’ (compare DL 7.72).
162 Sources for Two ØÆ æ
Ø in the Categories
 Stoic categorical propositions are wrongly omitted, such as ˜ ø
 æØÆ E. (Dexippus refers to this example as ‘composite’, but
on the Stoic view such a proposition was in fact ‘simple’. Com-
pare DL 7.70, where ˜ ø  æØÆ E is an example of a Stoic
simple ‘categorical’ proposition, and see Chrysippus at 7.65.)
 The categories Ø E and 
å Ø are wrongly separated, and
should be classed together under Œ Å
Ø. (Compare the three
groups of critics of the Categories at Simplicius 63,4–66,41; these
two categories are also called by Aristotle ØF and 
å , as
at An. Pr. 83a22; and the Stoic IæåÆ also have these names, as
DL 7.134 informs us.)
 Metaphorical and figurative expressions are wrongly omitted,
such as ‘I åÆ Ø
’ or ‘Æ Æ   ”Å’. (Compare Por-
phyry 59,5–14 and Simplicius 18,27 for the shared criticism by
Athenodorus and Cornutus that the Categories omits ‘figurative’
expressions.)
 Modal expressions are wrongly omitted, such as e K I ªŒÅ
and K å
ŁÆØ and æå Ø .
 Quantifiers such as e A are wrongly omitted.
These considerations, and especially the first set of criticisms
regarding conjunctions and articles, roughly correspond to Simplicius
64,20–65,12. The second natural division of Dexippus’ treatment
(Part II above), which overlaps with the first, is also loosely affiliated
with the following section of Simplicius, from 65,13 to 66,31.
 The ten categories fail to comprehend a number of (chiefly
logical and metaphysical) entities, including: being, the one,
plurality, non-being, ‘is’, same, other, definition, genus, pro-
prium, whole and part, and the overarching (Stoic) category Ø.

Summary of criticisms of the ‘division’ of ten categories

Combining the sources that we have, we are able to piece together a
broad picture of the combined criticisms of Athenodorus and Cor-
nutus against the division of the ten categories. Through a close
examination of the sources, we may also be able to distinguish
Athenodorus from Cornutus in the tradition.
Porphyry (58,3–59,14) informs us that ‘Athenodorus and Cornutus’
treated the
Œ of the Categories as ‘expressions qua expressions’,
 Stoic Critique 163
and sought to demonstrate that it failed in comprehending this sub-
ject. Porphyry also tells us (86,20) that their objections were chiefly
advanced in three works: Athenodorus’ Against the Categories of
Aristotle and Cornutus’ Arts of Rhetoric and Reply to Athenodorus,
and Porphyry’s compressed presentation may be taken to suggest
that all three of the groups of criticism—excess, deficiency, and
substitution—are to be found in these two Stoics. Simplicius (18,26)
reiterates Porphyry’s assessment, and explicitly distinguishes Atheno-
dorus’ Against the Categories, which criticized only the number or
division of categories, from Cornutus’ Reply, which bore a greater
resemblance to the wide-ranging (æe a ¼ººÆ  Æ
å  ) cri-
tiques of Lucius and Nicostratus (62,24–63,3). Simplicius distin-
guishes Athenodorus’ specific criticism about the number of
categories (based on various excluded º Ø), on the one hand, from
the other criticisms of the text, such as its thematic consistency, which
might be separately ascribed to Cornutus.
Dexippus then provides specific samples of criticism. I think it is
possible that the first long section of criticisms and responses,
expressed as a single disquisition by Dexippus and described as Part
I above, corresponds to Athenodorus’ criticism of the ‘number’ of ten
categories and its inability to comprehend º Ø, and so attracts a
defence founded chiefly on the distinction of the ‘primary’ and
‘secondary’ imposition of º Ø. Conversely, I think it possible that
the second series of criticisms and responses, expressed as a question
and answer dialogue between Dexippus and Seleucus and described
as Part II above, may be part of Cornutus’ wider critique, which also
includes several factors discussed below under other sections.
The reasoning behind these criticisms is associated with a number
of Stoic doctrines. The specific arguments in Dexippus make clear
that a strong initial motivation for the criticism of the Categories—
presumably as undertaken by Athenodorus—was the incompatibility
of Aristotle’s notions of predication with those espoused by the Stoics
following Chrysippus. Thus the Stoics perhaps found that the Cat-
egories omitted hypothetical (DL 7.69) and disjunctive (7.72) pro-
positions as well as what their theory regarded as ‘categorical’
propositions (7.70). More specifically, as I suggest above with refer-
ence to Barnes,59 later Stoics such as Cornutus may have found

59
‘Aristotle and Stoic Logic’, 41–3.
164 Sources for Two ØÆ æ
Ø in the Categories
particular fault with the ‘simple’ or atomic units of the Categories
being terms rather than sentences or propositions.60
Where Porphyry, Dexippus, and Simplicius feel compelled to focus
is less on the particulars than on the broad distinction of the so-called
‘primary’ imposition of expressions (qua significant of beings) from
the ‘secondary’ imposition of expressions as linguistic elements, in
order to associate the Categories with the former. It is important to
the defenders of the Categories to establish this distinction, and they
appear to suggest that the Stoic critics of the Categories erred in failing
to recognize it.
But as noted above in our initial treatment of Cornutus, this
language and distinction, as wielded by Porphyry and his followers,
is itself Stoic, answering to the Stoic distinction of dialectic into the
º Œ  or ‘what is signified’ from the class of studies concerned with
expressions or—for lack of a better word—with grammar.61 On the
orthodox Stoic view, evidently pioneered or at least elaborated by
Chrysippus, the signified º Œ  are separate from verbal expressions
or º Ø as such, which mirror the º Œ  (see DL 7.58). The ‘primary’
impositions with which Porphyry claims the Categories to be con-
cerned might well be interpreted in a Stoic framework as the º Œ ,
while the ‘secondary’ impositions of expression would be just the
verbal expressions that signify them.62
If the Stoics following Chrysippus made such a distinction as
Porphyry seems to claim they neglected, it seems likely that either
(a) Athenodorus ignored that distinction and ascribed a purely ‘ver-
bal’ skopos to the Categories for polemical reasons, or (b) Atheno-
dorus found reason in the Categories for interpreting the text within a
purely linguistic scope. Indeed, some modern commentators have
adopted the same approach.63 As we find later in the tradition (e.g.
Ammonius in Cat. 9,1 and following), those who argued that the
Categories concerned words alone relied on Cat. 1a16, ‘of things that
are said . . . ’ It is also possible that Athenodorus was drawn to this

60
On this, see also M. Frede ‘Stoic vs. Aristotelian Syllogistic’, in Frede 1987.
61
See Frede 1987, ‘Principles of Stoic Grammar’: 30 and following.
62
For the Stoic position itself, see for example Barnes (1993), Allen (2005), Long
(2005), Ebbesen (2005).
63
Consider for example the difficulty of determining whether paronyms are
entirely linguistic items, or non-linguistic, mixed-level items (Hintikka 1959: 141)
whose relationship is described linguistically (Annas 1974: 151).
 Stoic Critique 165
view by the apparently linguistic nature of the onymies; in the next
section I develop this suggestion further.

Athenodorus and Cornutus on the ‘Division’ of Chapters

As I noted above, there are actually three species of ‘division’ at play
in the single passage where Simplicius introduces the problem of
ØÆ æ
Ø (in Cat 18,22–19,7): (1) the ØÆ æ
Ø of chapter headings
in the Categories, a problem shared with the Alexandrian commen-
tators and associated with Stoic criticisms against the structural
coherence of the text of the Categories; (2) the ØÆ æ
Ø of the ten
categories and their inability to comprehend all lexeis, drawn from
Porphyry, and associated with Athenodorus’ and Cornutus’ distinct-
ive position about the skopos of the work; (3) the internal ØÆ æ
Ø of
the onymies and the question whether it is a division of names or of
things, which is incidentally relevant to the preceding two problems.
In this section, I will attempt to distinguish the sources of (1) and (2).
When Simplicius reports on ‘the book’s division into chapter
headings’ ( N a Œ çºÆØÆ ØÆØæ
ø F غ ı, 18,22), he remarks
that some commentators have failed to notice the ‘division [of the
chapters] according to their natural joints’ (ŒÆ ’ ¼æŁæÆ ØfiæÅ ÆØ,
18,23) and therefore they suppose that the Categories is organized
haphazardly like a hypomnematic work (18,22–6):
It remains for us to speak about the book’s division into chapters. Some
commentators, paying no attention to how the chapters have been
divided according to their natural joints and to how well they fulfil
the purpose of the overall subject, nor to how they maintain their
mutual continuity, consider that the chapters just lie there in the
haphazard manner typical of hypomnematic writings. (tr. Chase
2003)64
This remark apparently derives from a source common to the Alex-
andrian scholastic tradition on the ‘chapter headings’ of the Categor-
ies. Simplicius will return to that source at 19,9, where, in a passage

64
¸Øe b e  æd B N a Œ çºÆØÆ ØÆØæ
ø F غ ı ºª Ø
º  ÆØ, – Ø  c KØ

Æ  ‹ø ŒÆ ’ ¼æŁæÆ ØfiæÅ ÆØ ŒÆd ‹ø 
æe e ‹º
Œe åæ Æ IºÅæF
Ø ŒÆºH ŒÆd c æe ¼ººÅºÆ
ı å ØÆ
ØÆ
fiÇı
Ø ,
ØÅe Œ E
ŁÆØ  Çı
Ø a Œ çºÆØÆ (25) ŒÆ a e  ÅÆ ØŒe
æ (ed. Kalbfleisch 1907).
166 Sources for Two ØÆ æ
Ø in the Categories
echoed in Alexandrian commentators such as Ammonius, Philopo-
nus, and Olympiodorus,65 he describes the structural division as
‘tripartite’ (‰ N  ªºÆ æÅ Ø º E , æØ æb e غ  ).
However, Simplicius first diverges into a separate subject and,
perhaps, a different source. The divergence begins at 18,26 and is
marked by a shift in the reference of the word ‘division’ (ØÆ æ
Ø).
For context, here is the full text of 18,22–19,9 divided into four parts
(with translations from Chase 2003 and Greek text from Kalbfleisch
1907):
(i) (18,22–6) It remains for us to speak about the book’s division into
chapters. Some commentators, paying no attention to how the chapters
have been divided according to their articulations and to how well they
fulfil the purpose of the overall goal (
Œ), nor to how they maintain
their mutual continuity, consider that the chapters just lie there in the
haphazard manner typical of hypomnematic writings.66
(ii) (18,26–19,1) What is more, some contradict Aristotle and reject
his division: of these, some claim that it is uselessly redundant, others
that it omits many things, like Athenodorus and Cornutus, who believe
the goal (
Œ) is about expressions (º Ø) in so far as they are
expressions. They bring forward many expressions as examples, some
literal and others figurative, and thereby think they refute the division,
since it has not included all possible expressions.67
(iii) (19,1–7) These people also think that there is a division of
names into homonyms, synonyms, and paronyms, and suppose that
the book is a motley heap of logical, physical, ethical, and theological
speculations. For them, the speculations about homonyms, synonyms,
and paronyms are logical, as is, moreover, the one about opposites.
Those about movement, they say, are physical, while those about virtue

65
As found in Ammonius 14,3–15,2, Philoponus 13,6–33, Olympiodorus Proleg.
25,5–24, David (Elias) 134,29 and Arethas 138,21–139,12.
66
¸Øe b e  æd B N a Œ çºÆØÆ ØÆØæ
ø F غ ı ºª Ø
º  ÆØ, – Ø  c KØ

Æ  ‹ø ŒÆ ’ ¼æŁæÆ ØfiæÅ ÆØ ŒÆd ‹ø 
æe e ‹º
Œe åæ Æ IºÅæF
Ø ŒÆºH ŒÆd c æe ¼ººÅºÆ
ı å ØÆ
ØÆ
fiÇı
Ø ,
ØÅe Œ E
ŁÆØ  Çı
Ø a Œ çºÆØÆ ŒÆ a e  ÅÆ ØŒe
æ .
67
ŒÆd c ŒÆd I غªı
Ø ÆP fiH Ø  IŁ F  c ØÆ æ
Ø , ƒ b ‰
º  Çı
Æ  Å , ƒ b ‰ ººa Ææ E
Æ u
 æ ˚æ F  ŒÆd ŁÅ øæ,
¥ Ø   æd º ø N Ø e
Œe r ÆØ ŒÆŁe º Ø N
, ººa º Ø
溺  a b Œıæ Æ, a b æØŒ, Kºªå Ø Y ÆØ c ØÆ æ
Ø , ‰ P

Æ a º Ø  æغÆF
Æ ·
 Stoic Critique 167
and vice are ethical, just as the philosophical considerations about the
ten genera are theological.68
(iv) (19,7–9) In fact, however, the truth is otherwise. [Aristotle] is
not carrying out a division (ØÆ æ
Ø) of names [in the onymies], for if
he were, he would not have omitted heteronyms or polyonyms. If we are
to divide it into large parts, the book is tripartite . . . 69
For the surrounding passages (i) (18,22–6) and (iv) (19,9) and fol-
lowing, the ‘division’ is of chapters ( N a Œ çºÆØÆ ØÆØæ
ø) or of
major headings ( N  ªºÆ æÅ Ø º E ). But during the intervening
passage (ii), beginning from ήd c ήd at 18,26, where Simplicius
introduces Athenodorus and Cornutus, he turns to discuss the div-
ision of the actual categories, not the chapters, with specific reference
to the subject (
Œ) of the work. This intervening passage seems to
draw upon Porphyry, as I noted above by comparing the Question
and Answer commentary directly. At (iii) 19,1, however, the corres-
pondence with Porphyry’s commentary By Question and Answer
ends as abruptly as it began. Whereas Porphyry proceeds to describe
the views of Herminus on the skopos of the Categories, Simplicius
instead returns to the structural coherence of the text and the ony-
mies in particular (see above). This is a different problem and a
separate line of criticism than the contention that the Categories has
omitted certain kinds of º Ø. Porphyry’s smaller commentary also
covers the affiliation of the onymies to the rest of the work from 59,34
onward. But Porphyry treats the topic quite differently than does
Simplicius’ source for (iii) Simplicius’ source discusses the onymies in
the context of a division of the Categories by its ‘joints’—namely, the
Stoicizing organization of philosophy as an animal into the discip-
lines of ethics, logic, and physics.70 This source is interested in the
Stoic critics’ views on the relationship of the onymies (which are,
according to the Stoics, ‘logical’) with the rest of the work (which is, in

68
Q ŒÆd ØÆ æ
Ø H O  ø Y ÆØ Ø E
ŁÆØ N › ıÆ ŒÆd
ı  ıÆ ŒÆd
Ææ ıÆ ŒÆd r ÆØ e غ  Æ ÆH Ł øæÅ ø
øæ Æ ºÆ ı
Ø
ºªØŒH ŒÆd çı
ØŒH ŒÆd MŁØŒH ŒÆd Ł ºªØŒH · r ÆØ ªaæ a b  æd ›ø ø
ήd
ı ø ø ŒÆd Ææø ø
ŒÆ Æ ºªØŒ,  Ø b ŒÆd e  æd H I ØŒ Ø ø ,
a b  æd ŒØ 
ø çı
ØŒ, MŁØŒa b a  æd Iæ B ŒÆd ŒÆŒ Æ, u
 æ Ł ºªØŒa a
 æd H ŒÆ ª H çغ
çÆ Æ.
69
e b IºÅŁb På o ø å Ø· h ªaæ ØÆ æ
Ø Ø E ÆØ H O  ø · P ªaæ i
a  æ ıÆ ŒÆd ºı ıÆ ÆæBŒ . Iºº’ 
Ø , ‰ N  ªºÆ æÅ Ø º E , æØ æb
e غ  .
70
DL 7.39–40.
168 Sources for Two ØÆ æ
Ø in the Categories
their view, ‘ethical’, ‘physical’, and ‘theological’). For Simplicius’
source, the correct reply to the Stoics is that the first chapter of the
Categories is not concerned with a ‘logical’ or linguistic subject
(19,7–9 = iv above).
But this is not Porphyry’s reply. Porphyry’s response to the prob-
lem of the onymies, as given in the QA commentary, is that the
onymies are a necessary propaedeutic for the rest of the work
(Porph. 60,1 and following). In fact, Simplicius cites that very
response of Porphyry later in his text (19,27 and following, where
the questioners are ‘Lucius and Nicostratus’).
Passage (ii = 18,26–19,1) above, I think, draws from Porphyry,
directly or indirectly: but the passages before and after it derive
directly from a different source. I believe this source is Syrianus or
Iamblichus. The source contends that the ‘onymies’ do not concern a
linguistic subject, because they omit heteronyms and polyonyms.
The argument that polyonyms and heteronyms belong to ‘linguistic’
study is directly associated with Syrianus in a later passage (23,13–15:
æe º Œ ØŒc <Aºº > ÆæÆ
Œ ıc ›æA j æe ÆP c c H
æƪ ø K º łØ ). Thus it appears likely that Syrianus is the
immediate source of the argument at (iv) 19,7–9. Moreover, Simpli-
cius’ citation of Syrianus distils ideas about the status and scope of the
‘onymies’ which are ascribed by name to Iamblichus (e.g., 23,25–34,5
and earlier at 22,1–2). This association seems well in keeping with
Simplicius’ own statement of intent at 3,9, where he remarks that he
is reducing Iamblichus much as Syrianus did, but less radically;
and also with Simplicius’ comment at 13,17 that Syrianus ‘clarified’
(
ÆçÅ Ç Ø) Iamblichus on the true subject of the Categories, namely
that the work treats significant words insofar as they are significant of
beings.
For reasons outlined below, I believe it is important to pry these
sources apart in the Simplician text. Simplicius’ references are divided
as follows (again using the numeration above, for the first three
items):
 (i) 18,23: Ø  question the division of the chapter headings.
 (ii) 18,26: What is more (ŒÆd c ŒÆd) Ø  speak against the
division of the categories. There are two groups, mirroring two
of the three groups at 63,4–66,31:
 18,27: Some (ƒ  ) claim the division is excessive
 Stoic Critique 169
 18,28: Others (ƒ ) such as Athenodorus and Cornutus hold
that the division omits many lexeis, as the ten categories concern
‘verbal expressions’
 (iii) 18,31: ‘Who also’ (Q ŒÆd) criticize the internal division of the
onymies and the disciplinary division of the work as a whole, as
the onymies concern a linguistic subject, while the ten categories
concern theology, etc.
The criticisms contained in this passage, taken together, appear to
be inconsistent. The central lines (ii) contain criticisms explicitly
ascribed to ‘Athenodorus and Cornutus’, who hold that the
Œ
of the Categories is ‘verbal expressions insofar as they are verbal
expressions’, divided into ten categories. But the critics represented
in the following part of the passage (iii–iv) complain that the ten
categories are theological speculations that confusedly follow the
‘linguistic’ onymies; this can hardly be consistent with the position
of ‘Athenodorus and Cornutus’ that the
Œ is ‘verbal expres-
sions’. For Athenodorus and Cornutus, the Categories—and the cen-
tral list of the ten categories in particular—have a clear linguistic

Œ; for the critics of the chapter division, the ten categories are
‘theological’, and the work itself has no single coherent
Πat all,
treading between logic, physics, ethics, and theology.
To summarize: the critics of the chapter division, represented in
(i) and (iii–iv), may not be identical with Athenodorus and Cornutus,
represented in (ii). Indeed, the passage ending at 18,26 could flow
directly into 19,1 if the intervening passage (ii) were removed. The
passages 18,22–6 and 19,1–8 treat the same subject (the structural
coherence of the work) and are consistent with one another in their
use of the term ØÆ æ
Ø. The intervening passage is associated with
criticisms about the ‘division’ of the ten categories, where Simplicius
is primarily drawing on Porphyry for his response; but the surround-
ing passages are associated with criticisms about the chapter division,
where Simplicius is primarily drawing on Syrianus and Iamblichus
for his response.
So who were the critics of the chapter division in (i) and (iii–iv)?
After all, the criticism of the Categories’ chapter division as a blended
‘heap’ of philosophical disciplines is not a characteristically Stoic line
of attack, at least judging from one report of school orthodoxy. For
the Stoics, according to Diogenes Laertius, the disciplines of logic,
physics, and ethics are ‘not independent . . . but all blended together
170 Sources for Two ØÆ æ
Ø in the Categories
(  åŁÆØ ÆP )’ in philosophical discourse; indeed, ‘it was not usual
to teach them separately’ (7.40). Thus it is not especially obvious that
a Stoically-inclined philosopher would, on doctrinal grounds alone,
object to a text that mixed philosophical subjects in this way. We can
establish with confidence that Athenodorus made no such criticism,
as Simplicius elsewhere informs us (62,25) that Athenodorus criti-
cized only the division of the ten categories and seemingly had
nothing to do with the criticism of the structural coherence of the
text. But Simplicius clearly sees enough connection—aside from the
common word ØÆ æ
Ø—to link the lines and senses of 18,22–19,9
together. We do know that Cornutus’ criticisms were broader-
ranging in his Against Athenodorus and Aristotle and his Rhetoric.
Conceivably Cornutus alone criticized the chapter division, and
the internal inconsistency of the argumentation in our passage is to
be explained by a difference of opinion between Athenodorus and
Cornutus—for we know that they disagreed, although we do not
know what they disagreed about.
Indeed, there is some evidence that Cornutus, in contrast to Athe-
nodorus, understood the ten categories as non-linguistic items. (He
may, of course, have presented this interpretation as a mere side effect
of the text’s poor organization or the author’s inability to distinguish
consistently between use and mention, or between grammatical,
dialectical, and metaphysical subject matter.) Simplicius directly
compares the views of Athenodorus and Cornutus on the subject of
what Aristotle meant by ‘relative’ (in Cat. 187,24–34). For Cornutus,
it is appropriate to understand the Aristotelian relative in terms of
its signification of reality (æ Ø r Æ çÅ
Ø x 
ıæ
  Ø
æe  æ 
å
Ø, P  Ø 
ı ÆŒ ØŒ . . . Iºº  æO

Æ
Ø . . . ); the language suggests that the Relative is interpreted
as a º Œ  , directly significant of reality, rather than a ºØ-in-itself,
that is, a linguistic item that is several steps removed from reality. As
discussed below, this passage suggests that Cornutus holds a less
strictly ‘linguistic’ view about the subject matter of Aristotle’s ten
categories than does Athenodorus. Indeed, if Cornutus shared the
position, presumably Athenodorus’, that the sole subject of the work
was º Ø ŒÆŁe º Ø, it would be difficult for him to argue coher-
ently that the work also transgressed disciplinary boundaries between
physics, ethics, and theology—that critique would seem to be at odds
with the supposition that it was strictly linguistic. Perhaps, as I would
suggest, the later commentators’ attribution to both Athenodorus and
 Stoic Critique 171
Cornutus of an overly linguistic reading of the text should stick
primarily to Athenodorus, and might even rely partly on Cornutus
himself as an intermediary who entertained such criticisms.
Such an interpretation could be coupled with the evidence dis-
cussed earlier that the initial response to Athenodorus’ criticisms was
constructed within a purely Stoic framework—in which Atheno-
dorus’ (presumably intentional) error was to describe the ten categor-
ies without regard for the primary part of dialectic. Given this, we
might suggest that Cornutus, in his Against Athenodorus, promoted
the argument (against Athenodorus) that the ten categories involved
lekta rather than º Ø ŒÆŁe º Ø, but simultaneously promoted the
argument (against Aristotle) that the Categories was accordingly
incoherent, crossing boundaries between logic (in the onymies),
theology (in the portion of the work where claims are made about
the nature of ousia), ethics, and physics. This would account for the
discontinuity of 18,22–19,9, with its two senses of ØÆ æ
Ø, as well as
Simplicius’ choice to combine them into a single report of the views
maintained by ‘Athenodorus and Cornutus’.71

III. CONCLUSIONS

Porphyry (and Simplicius following Porphyry) imply that Atheno-

dorus and Cornutus criticize the Categories because they do not
recognize the distinction of the primary and secondary imposition
of language on realities. But this distinction is traditional in the Stoic
bipartition of dialectic. The historical figures referred to as ‘Atheno-
dorus’ and ‘Cornutus’ would have written commentaries that were
sensitive to this distinction.
Athenodorus, perhaps, identified the subject matter of the ‘ten
categories’ with the second (linguistic) component of Stoic dialectic.
He contended that the onymies were solely linguistic—as have some
modern critics—and evidently founded a criticism of the Categories
as linguistic (and so defective) upon this exegesis. In this connection,

71
For reasons of space, I omit here other challenges posed by Athenodorus and
Cornutus to the characterization of specific categories; for these, see for example my
discussion in Griffin 2009: ch. 6, and the surveys referenced in the opening footnotes
to this chapter.
172 Conclusions
I think that we should assign to Athenodorus and Cornutus the
counter-example of conjunctions ascribed to Lucius at Simplicius in
Cat. 64,18–65,13—in light of doctrinal inconsistencies with the views
of Lucius expressed elsewhere, coupled with the lacuna in Simplicius
and the appearance of a similar argument under the ‘Stoic’ banner in
Dexippus. I also suggest that it is unlikely that ‘the Stoics’ at
66,32–67,9 are directly related to Athenodorus and Cornutus, and
they are not found in Simplicius’ major threefold partition of critics of
the categories in 62,25–31. This is significant, because the division of
four ‘categories’ attributed to the Stoics there may have had nothing
to do with Athenodorus’ and Cornutus’ criticisms: we need not
suppose that they were motivated by a desire to substitute a fourfold
Stoic scheme of categories for Aristotle’s. Moreover, I have suggested
that it might be reasonable to locate Athenodorus’ critical activity
before Andronicus and the burst of attention drawn to the Categories
over the first century; perhaps Athenodorus encountered the text in
its guise as a proem to the Topics introducing simple terms as such,
since his interpretation would miss the mark somewhat against an
Andronican or Boethan exegesis of the text as ŒÆ Ūæ ÆØ.
There also appear to have been differences between the criticisms
of Athenodorus and Cornutus, which were sometimes at odds. This is
evident not only from the title of Cornutus’ book Against Atheno-
dorus—or Reply to Athenodorus—but also from the fact that the
arguments contained in 18,22–19,9 (in the numbering above, [i]
and [iii] vs. [ii]) are incompatible. Therefore, I have contended that
this well-known passage should be picked apart based on its multiple
incompatible uses of ØÆ æ
Ø, one use referring to the division of ten
categories, the others to the chapter division of the work; and
I propose that Athenodorus criticized only the division of ten cat-
egories (thus focusing on the ‘lexical’ omissions of the text, cf.
Simp. 62,25), whereas Cornutus amended this criticism, suggesting
various cases where the Categories appeared to deal with º Œ  rather
than º Ø-as-such. Perhaps it is reasonable to locate Cornutus’
activity after Andronicus and Boethus, taking into account their
more sophisticated, semantical interpretation of the treatise. But
Cornutus also appears to have challenged the overall literary coher-
ence of the work based on this interpretation. There is some (limited)
evidence for this in Cornutus’ rebuttal of Athenodorus’ exegesis of
certain categories.
 Stoic Critique 173
Porphyry’s defence of the Categories may also have some Stoic
roots with Cornutus. Porphyry treats its subject matter as (in effect)
º Œ  rather than º Ø ŒÆŁe º Ø, and I have suggested, tentatively,
that Cornutus’ own objections to Athenodorus may be a source of
that defence, based on the inconsistencies in 18,22–19,9. We may also
find in the testimony of Dexippus, and the identification of our
Athenodorus with the Athenodorus of DL 7.68, evidence for a sug-
gestion previously made by Barnes that it was the basic distinction
between simple term logic and simple propositional logic that motiv-
ated some Stoic criticism of the Categories, at least in the Imperial
period: in brief, the Categories account of ‘things said without com-
bination’ may have been viewed (at least from a purely polemical
standpoint) as somehow contradictory of the Stoic doctrine of ‘simple
assertions’ (cf. Dexippus in Cat. 12,3–11).
 Part C

Peripatetic Synthesis and Response

 6

Boethus of Sidon

Some commentators, however, also applied deeper thoughts to

the work, as did the admirable Boethus.1
Simplicius, in Cat. 1,8 (tr. Chase 2003)

To any reader of Simplicius’ commentary on the Categories, Boethus

of Sidon2 appears to be a dominant figure, whose intellectual vigour
left his interlocutors in the dust (11,23). Outside of the Neoplatonic
tradition, however, the work of Boethus is scarcely attested in Greek.
He has left little or no explicit record in Latin.3 In coming to grips
with his life and thought, we are limited to deductions that may be
drawn from later doxographies, and to a reference in Strabo’s Geog-
raphy. Despite all of this, he was arguably one of the first-rank
thinkers of his age (Rashed 2013: 53).

1

Øb 
Ø ŒÆd ÆŁı
æÆØ æd ÆP
e ØÆ ÆØ ŒÆ
åæ Æ
, u æ › ŁÆıØ
BÅŁ (ed. Kalbfleisch 1907).
2
On Boethus see recently Rashed 2013 and Chiaradonna 2013. Classic surveys
include P. Moraux, Der Aristotelismus bei den Griechen, I (Berlin, 1973) 143–79;
H. B. Gottschalk, ‘Aristotelian Philosophy in the Roman World from the Time of
Cicero to the End of the Second Century ad’, ANRW II.36.2 (1987), 1107–10,
1116–19; and J-P. Schneider, ‘Boéthos de Sidon’, in Goulet (1994) vol. 2, 126–30.
See also E. Zeller, Die Philosophie der Griechen in ihrer Geschichtlichen Entwicklung,
T. III: Die Nacharistotelischeanni Philosophie (Fotomech, Nachdr. der 5. Aufl., Leipzig
1923), 1, 642 and following; K. O. Brink, ‘Peripatos’, Paulys Realencyclopädie, suppl. 7
(1940), cols. 899–949; F. Wehrli, Die Schule des Aristoteles (2nd ed. Basel, 1967–69);
R. W. Sharples, ‘The Peripatetic School’, in D. J. Furley (ed.), Routledge History of
Philosophy, vol. 2: From Aristotle to Augustine (London, 1997), 147–87; and
L. Tarán’s helpful review of Moraux, ‘Aristotelianism in the 1st century bc’, in
Taran 1981 = L. Tarán, Collected Papers (Leiden, 2001), 479–524.
3
While citations of Boethus Stoicus are common in the Latin corpus, there is no
certain citation of Boethus Peripateticus in Latin—unless Macrobius in Somn. Scip.
1.14.19 refers to Boethus Peripateticus; the reference is ambiguous.
178 Boethus of Sidon
In this chapter, I aim to sketch Boethus’ responses to criticisms of
the Categories, and his elaboration of Andronicus’ interpretation of
the treatise, with important additions and revisions. I have already
argued (ch. 2) that Boethus valued the Categories as an introductory
treatise, furnishing  ªæÆçÆ of primary genera to untutored stu-
dents before they have mastered technical definitions; I also argued in
ch. 2 that Andronicus had taken a similar view, and that Boethus
followed him. I have also argued that Lucius (ch. 4) attacked the
Categories as an inadequate work of ontology, due to its omission of
any account of intelligible being, while Athenodorus (ch. 5) criticized
the treatise as an inadequate work of rhetorical or grammatical
theory, due to its omission of many kinds of speech. Boethus, in
response, maintained that the Categories concerns significant words
just qua significant of beings: this is what restricts its scope in
ontology to sensible beings, the referents of ordinary speech, and
excludes intelligibles (see Simplicius in Cat. 78,5 for Boethus’ pos-
ition, and compare 74,3–17, 82,4–6).4 He restricts its scope in gram-
mar and rhetoric to significant terms. (While this formula is broadly
in agreement with the view that I attributed to Andronicus in ch. 2,
Boethus also goes well beyond Andronicus, in adding a role for
mediating concepts between words and beings. Here perhaps he
was influenced by the first chapter of De Interpretatione, whose
Aristotelian authorship had been rejected by Andronicus in the pre-
ceding generation, but would go on to a very distinguished career of
influence in later philosophy.)
Treating the Categories as in part a work of ontology led Boethus to
problematize the relationship between the Categories and the central

4
Ben Morison helpfully points out to me that this is an odd position for a reader of
the Categories to adopt, given that Aristotle certainly does refer to non-perceptible
items such as the soul in the treatise (for example, in ch. 2, 1a20–b9), to say nothing of
time and number. Perhaps Simplicius could reply that by ‘soul’ Aristotle means the
enmattered form of an animal, which is available to perception and ordinary language,
and that the ‘time’ and ‘number’ under consideration here are again enmattered
quantities that we can perceive. Boethus seems explicit, at Simplic. in Cat. 78,5, that
the Cat. is not concerned with intelligible P Æ; another passage (in Cat. 65,20–1)
implies that Boethus wishes to distinguish clearly between intelligible, substantial
number, on the one hand, and the sensible quantity, so perhaps he would have taken a
line similar to Simplicius’, namely, that the Categories addresses itself to the latter and
not to the former. But in any case, the Neoplatonist commentators, and perhaps
Boethus, seem to have found the sensible examples of primary substances (like
Socrates and Bucephalus) strongly suggestive that Aristotle wrote Cat. to talk about
sensible substances.
 Boethus of Sidon 179
books of the Metaphysics, a development which would contribute
to influential ontological interpretations of the Categories that
have persisted until today. By restricting its subject or Π and
emphasizing that the scope of the treatise was limited to per-
ceptible things, he also laid the groundwork for the exegesis of the
text later adopted by Porphyry, which made it compatible with
Platonism.

I. LIFE

Strabo (63/64 bce–c.24 ce), writing in the early first century ce,
reports that he ‘studied Aristotle’ with Boethus (ı çغç Æ 
 E
a æØ

º ØÆ, Geog. 16.2.24). The word ı çغç Æ 
may imply that Strabo was taught by Boethus, or that they read
Aristotle as fellow-students of a common master.5 As Strabo’s birth
can be fixed,6 Boethus’ activity is generally assumed, chiefly on the
authority of this passage, to have fallen early in the first century bce
(if he was Strabo’s tutor), or later in the first century bce (if they were
contemporaries). The latter deduction seems more plausible, for
reasons I suggest below. Some further information may be gleaned
from Geog. 16.2.24: Strabo reveals that Boethus had a brother
(I ºçe ÆP
F) named Diodotus, and proceeds to mention two
‘contemporary’ (ŒÆŁ’ A) Sidonian philosophers, Antipater of
Tyre and Apollonius, the latter flourishing ‘a little before my time’
(ØŒæe æe H). Both references confirm the approximate chron-
ology of the first century. They may also suggest some Stoic associ-
ations.7 I assume here that Boethus is a Peripatetic from Sidon who is

5
Perhaps under Andronicus of Rhodes (cf. Strabo, Geog. 14.2.13), or under
Xenarchus of Seleuceia (cf. 14.5.4). For literature on the question, see the earlier
chapter on Andronicus.
6
On Strabo’s dates, see W. Aly, ‘Strabo (3)’, Realencyklopädie der classichsen
Altertumswissenschaft 4 (1932), 76–155.
7
There were two Tyrian Antipaters who may answer to this reference, both Stoics,
the younger of whom died just before 45 bce (cf. Cicero, de Off. 2.24). This Apollo-
nius, according to Strabo, ‘published a tabulated account ( ÆŒÆ KŒŁ d) of the
philosophers of the school of Zeno, and their books’. Together with Boethus’ own
namesake, the Stoic ‘Boethus of Sidon’ (discussed below), there was a clear Stoic trend
in recorded Sidonian philosophy.
180 Life
not to be identified with a Stoic from Sidon of the same name (see
note in Appendix 1).8
From Simplicius, we may be able to hazard a guess that Boethus
was active into the middle of the first century ce, and is therefore
more likely to have been Strabo’s fellow-learner than his mentor. For
Boethus appears to have responded to certain criticisms of the Cat-
egories, many of which are ascribed by Simplicius to the critics Lucius
and Nicostratus (e.g. Simplicius in Cat. 50,2, comparing 48,1; and see
ch. 4 on Lucius and Nicostratus for discussion).

Relationship to ‘Lucius and Nicostratus’

Each of Simplicius’ eight references to Lucius and his ‘school’ is
followed by a reply from Boethus (see ch. 4), implying that Lucius
published the relevant criticism of the Categories before or during the
life of Boethus. But the only reasonably firm point in this chronology
is Nicostratus, for whom Praechter (1922) has persuasively suggested
a date in the middle of the second century ce.9 Working backwards,
we are told by Simplicius that Nicostratus ‘came after’ Lucius (
’
ÆP
e, 1,19) and ‘took over’ his ideas (
a
F ¸ıŒ ı  ƺº ,
1,19–20). Perhaps there is enough force in Simplicius’ language to
suggest that, in the view of his own sources (with Porphyry the
nearest detectable authority), the lives of Lucius and Nicostratus
overlapped, or at any rate that they were treated as part of the same
circle. If Lucius was active sufficiently late that Nicostratus knew him,
and yet Boethus responded to Lucius’ aporiai, Boethus too must have
been active into the middle of the first century or so.
This is shaky ground. As we shall find, such ‘conversations’ in
Simplicius cannot always be taken as representative of historical
discussions;10 it seems more likely that Boethus presented certain
views that were later reinterpreted, for instance by the later second-
century Peripatetic Herminus or by Porphyry, as useful positions to

8
I am grateful to Stephen Menn for conversations that have persuaded me, since
2009, that Boethus the Peripatetic is also the likely target of Porphyry’s treatise Against
Boethus.
9
Our Nicostratus, on Praechter’s argument, was honoured by an inscription at
Delphi (Syll. 3 868) in the middle of the second century ce, late in his life.
10
See in particular the ‘table of responses’ in my chapter on Lucius and Nicos-
tratus, and discussion of dates below.
 Boethus of Sidon 181
maintain against later critics.11 But as I argued in ch. 4, we can claim
with some confidence that Boethus did write a book against the views
that Simplicius ascribes to Lucius and his like-minded associates (ƒ
æd
e ¸ŒØ), and that the source of these views may have been a
contemporary of Boethus, and certainly was not a contemporary of
Nicostratus.12
Thus we might take the view that Lucius was still active in the latter
half of the first century ce. If this is correct, then Boethus of Sidon,
who answered him, must have lived well into the first century ce—
long enough to spar with Lucius, and answer several of Lucius’
criticisms of the Categories. This line of reasoning makes it unlikely
that Boethus was significantly older than Strabo, and more probable
that the two men studied together as fellow-students, dating the birth
of Boethus to c.60 or 50 bce.

Relationship to Andronicus
Boethus may have studied under Andronicus, as Philoponus says (in
Cat. 5,16–19). Conversely, Boethus may have been a contemporary
and intellectual equal of Andronicus, as Düring, following Brink,
proposed.13 Boethus may also have taken up the scholarchate of the
Aristotelian ‘school’, as the tradition held (Ammonius, in An. Pr.
31,11). The sixth-century testimonies can all be questioned, but there
seem to be other reasons to treat Boethus as a follower of Andronicus.
I have put some weight on Aspasius (in Eth. 44,24), who has Boethus
cite and defend an adapted version of Andronicus’ account of passion

11
Our trust in the testimony of Simplicius for chronology may also be troubled by
his habit of implying, if not explicitly stating, that Boethus energetically rebutted the
problems raised by Nicostratus himself (58,14–28). But as I argue in the preceding
chapter, the only place this occurs is more likely to be an oblique response to ‘the
Lucians’, and in general I believe it is clear that Boethus is only represented by
Simplicius as replying directly to Lucius’ aporiai, or rather, to difficulties presented
by ƒ æd
e ¸ŒØ. Nevertheless it is certainly true, as Gottschalk (1990) writes,
‘Many of the criticisms [Boethus] refuted were repeated by later writers, including
Plotinus, to be refuted again by others reusing many of Boethus’ arguments’ (74).
12
Even if Lucius is in some respects a mere construct of Simplicius’ text, I will
suggest that the Lucius source was equally a construct of a commentary by Boethus
(that is, even if we cannot get to an historical Lucius, we can get to the Lucius that
Boethus set out as an interlocutor, and this figure was not a construction of the later
commentary tradition).
13
Düring (1957), 420–5; for earlier views see C.O. Brink in RE suppl. 7 (1940), 938.
182 Life
verbatim against a criticism that was also used to attack Chrysip-
pus14—this certainly does not prove anything, but it lends circum-
stantial support to the biographical tradition.
The tradition that Andronicus taught Boethus would suggest that
their lives overlapped. (Proximity in time might also be loosely sup-
ported by the fact that we have no evidence for intermediaries between
them.) We might also speculate that Lucius’ life (if he was indeed a
biographical individual) overlapped with Boethus’ and with Nicostra-
tus’, if we can place such weight on Simplicius’ language of ‘following’
and ‘taking over’. We might even reconstruct a sequential chronology
like that sketched in the following table, in which Andronicus overlaps
with Boethus, Boethus with Lucius, and Lucius with Nicostratus:

100–50 bce 50–0 bce 0–50 ce 50–100 ce 100–150 ce

Andronicus

Boethus

Lucius

Nicostratus

This is quite a speculative picture, and we cannot place much weight

on it. But it seems to me as plausible as any other interpretation of the
scarce evidence. According to this scheme (which would imply that
each party grew rather old), Lucius commenced a programme of
criticism of the Categories early in his career, perhaps motivated in
part by its growing popularity in Platonist circles, and made a name
for himself in this way. Boethus may have countered these criticisms
towards the end of his own career. Such a chronology also allows for
the interpretation that Boethus’ early work on the Categories played a
role in popularizing the text in philosophical circles, perhaps espe-
cially in Athens, and instigated a reaction from some Platonists,
represented by Lucius as a young exegete who may (with Dillon)
have belonged to the ascendant Neo-Pythagorean wing of the
school.15 Lucius may have been active in the city of Athens, judging
from the fact that Nicostratus—perhaps a member of his scholarly

14
See the earlier ch. 2 on Andronicus of Rhodes.
15
As Dillon (1977: 344–5) proposes that Lucius was a Pythagorean ‘of the strict
observance’.
 Boethus of Sidon 183
circle—was active in Athens; it would be elegant, then, if Boethus
were a ‘grand old man’ of Athenian Peripateticism early in Lucius’
lifetime, defending his own seminal work on the Categories against a
younger critic in a magnum opus which would set the tone for
centuries of discussion.
Boethus likely maintained that the Categories was appropriate to
novices in logic, whose æºÅłØ of the genera of predication requires
articulation (since Herminus is following Boethus at Porph. in Cat.
59,17–33). He also maintained that the treatise offers helpful ‘delin-
eations’ or  ªæÆçÆ of the genera, sketches that are able to stir up
our concepts and begin to excavate them (Simplicius in Cat.
159,10–15). So far as this he followed Andronicus. He did not follow
Andronicus in taking logic to be the first step in the curriculum of
philosophy (see discussion in ch. 2); but since he took the Categories
to deal with sensible items, perhaps we could imagine that he allowed
for it to be read as a work of physics, before the systematic pursuit of
logic later in the curriculum.

II. WORKS

Referring to Boethus as an exegete (KŪ , 29,30), Simplicius

bestows high praise upon his interpretive depth (1,18) and acumen
(11,23, drawing upon Porphyry), suggesting that Boethus answered to
Simplicius’ own criteria (7,23–8) for a capable exegete of Aristotle by
combining a wide knowledge of the texts (
H Æ
ÆåF
fiH çغçø fi
ª ªæÆø  Øæ) with respectful impartiality (I ŒÆ
)
regarding the authority of Aristotle. (It is worth noting that Simpli-
cius’ sources may not always be unanimous on Boethus’ success as an
interpreter of Aristotle and as a philosopher, but the consensus from
Porphyry onward certainly seems positive.)16

16
Shortly after stating that ‘I refuse to call him ignorant, such a respected man’ (P
ªaæ c Y ØØ i ‹
Ø Mª Ø KººªØ Icæ ª ª, 159,14), Simplicius proceeds to
do exactly that (Iª E b Œ E, 163, 19–20), which might possibly suggest some
divergence among his sources. While it seems probable that Simplicius’ information
concerning Boethus derives from Porphyry’s commentary Ad Gedalium, Boethus is
sometimes (e.g. 339,19) closer to Iamblichus, and is allied closely with Iamblichus
where Porphyry’s view differs slightly (342,23); and there are various cases where
Porphyry replies directly to, or criticizes, a view attributed to Boethus (78,4). Huby
184 Works
There is not yet a comprehensive edition of Boethus’ fragments,17
which cover a wide range of topics.18 ‘His main work’, as Gottschalk
comments, ‘was a grand-scale commentary on Aristotle's writing on
categories, but he had an encompassing interest in all areas of phil-
osophy’.19 The perception that Boethus’ ‘Hauptwerk’ concerned the
Categories, however, is a natural consequence of the space accorded to
Boethus by Simplicius in his Categories commentary, and the mag-
nitude and influence of the later Categories commentaries themselves.
This may reflect the interests of Porphyry and Iamblichus, upon
whom Simplicius relies, rather than the primary focus of Boethus
himself. In fact, to take the sparse citations in sequence, the attention
of antiquity, or at least of the preserved sources, initially focuses upon
Boethus’ psychology, ethics, and occasionally syllogistic (second and
third centuries for example with Aspasius) and his views on time and
being (fourth century) before turning to positions that are explicitly
associated with the Categories in the sixth century and the defences
that he reportedly prepared against Lucius and Nicostratus. Thanks
to that latest ancient tradition, we possess more information about
Boethus’ work on the Categories than about any other subjects that he
may have treated during his philosophical career.20 In what follows,
I will focus just on Boethus’ views on the subject matter of the
Categories and his responses to figures like Lucius and Athenodorus,

(1981) notes that one possible fragment of Boethus overlaps significantly with Iam-
blichus in Simplicius’ text, and it is likely that Iamblichus also played a significant role
in Simplicius’ value judgements on Boethus.
17
Marwan Rashed, Riccardo Chiaradonna, and Philippe Hoffmann are currently
directing a collection of the fragments of Boethus of Sidon, embracing a wide range
of intellectual topics, many of which converge upon themes that are relevant to
his exegesis of the Categories.
18
In an earlier survey (Griffin 2009), I reviewed approximately fifty named
citations in Greek which correspond with reasonable certainty to the Peripatetic
Boethus of Sidon, roughly half of which derived from Simplicius in Cat., and
I included the treatment On the Category ‘When’ surviving in the fourteenth-century
MS Laurentianus 71 and ascribed by P. Huby (1981) to Boethus (for a cautionary note
on the attribution, see Rashed 2013: 71 n. 43).
19
‘Boethus (4)’, Der Neue Pauly: ‘Sein Hauptwerk war ein groß angelegter Komm.
zu Aristoteles' Kategorienschrift, aber sein Interesse umspannte alle Gebiete der
Philosophie.’
20
His fragments cross the boundaries of logic, ethics, and physics, and nearly all of
his fragments shed some light on his positions on the Categories; it is difficult, for
example, to rule out the possibility that his positions on time or movement derive
from his commentary on the category When or the argument that several categories
should be collapsed under the heading of Motion.
 Boethus of Sidon 185
omitting other areas of his thought (which I attempted to explore in
some more detail in Griffin 2009: ch. 5).
We might distinguish two streams of work on the Categories
by Boethus, and for convenience I will treat them in separation.
(1) There is a set of comments offered on a line-by-line (ή
a
ºØ) basis, which, as I argue in concluding this chapter, may have
provided the structural model for Porphyry’s Ad Gedalium and,
eventually, for Simplicius in Cat. (2) There is also a series of responses
to various aporiai raised against the Categories, which survives in
named and unnamed jousts by antilogia (see for instance Simplicius
in Cat. 63,2). As I suggest below, such a commentary might be visibly
echoed in Porphyry’s commentary By Question and Answer.

Evidence for Line-by-Line Study of the Categories

Simplicius observes that Boethus carried out a ‘word-by-word exe-
gesis’ of the Categories (KŪ  . . . ŒÆŁ’ Œ
Å ºØ, 29,30)
and composed a ‘whole book’ On the Relative (‹º غ  ªæłÆ
æd
F æ
Ø, Simplic. 163,6). This latter work presumably under-
lies seven explicit references to Boethus over forty pages of Simplicius’
own treatment of the Relative (159,14–202,1; see also Luna 2013).
The work On the Relative might have constituted a ‘monograph’ by
Boethus, given Simplicius’ description of a ‘whole book’ (‹º
غ ) devoted to the topic.21 However, Simplicius does not else-
where use the phrase ‹º غ  in the sense of an independent
‘monograph’, but rather to denote a sequentially numbered book
within a complete text. Earlier in the Categories commentary (2,4),
for instance, Simplicius describes Plotinus’ work On the Kinds of
Being (6.1–3) as existing ‘in three whole books’ (
æØd ‹ºØ غ Ø),
and he repeats the phrase in this unambiguous sense in other com-
mentaries, with the adjective ܼ drawing attention to the magnitude
of the scholarly undertaking involved.22 It is probable, therefore,
that Boethus’ ‘book’ on the Relative formed a portion of a larger
work, either his ‘word-by-word exegesis’ of the Categories or the

21
On the force of this phrase, see Gottschalk (1990: 74), Moraux I (147–64), Prantl I
(540 and following).
22
At de Caelo 366,25 Simplicius observes that ‘Alexander supposed the first whole
book [of the work] (
e æH
 ‹º غ ) to concern the whole cosmos’, and the
phrase recurs elsewhere in the same work (674,4).
186 Works
commentary per interrogationem et responsionem that I attempt to
distinguish below.
Moreover, Simplicius signals that it is a whole book in order to
emphasize, rhetorically, the relative size of the work;23 this suggests
that Boethus’ exegesis of the Relative was particularly elaborate in the
context of his commentary, and that some of his other discussions,
pertaining to certain other categories, were contained in less space.

Responses to Ipoqßai about the Categories

In addition to this treatment of the Categories ŒÆŁ’ Œ
Å ºØ,
Boethus comments specifically on a number of problems or aporiai
raised about the Categories (which may or may not represent a
separate publication). At least some of these aporiai were raised by
commentators such as Lucius, and perhaps by Andronicus himself.
That Boethus focused, in some one of his works, on addressing
aporiai regarding the Categories is suggested by the reportage in
Simplicius, where Boethus is often cited as defending the Categories
against varied criticisms. This hypothesis is also bolstered by a spe-
cific passage in Cat. 58,27–59,4, where Boethus is described as ‘giving
in’ to a certain problem (K f
B
fi I æ fiÆ) by amending the text of
the Categories—suggesting that he typically responded to aporiai by
staunchly defending the text. In addition, I describe below several
places in Porphyry’s commentary By Question and Answer where
large tracts of discussion, and particularly responses to aporiai that
bear the brand of Lucius, might be traced to Boethus. In fact, I will
suggest that Porphyry’s twin commentaries On the Categories—
namely, the commentary ή
a Ø ŒÆd I ŒæEØ24 and the
word-by-word treatment Ad Gedalium25—are modelled on one or
two analogous commentaries by Boethus.

23
So, too, he emphasizes the scope of the work contained in Plotinus 6.1–3 by the
phrase
æØd ‹ºØ غ Ø: see above.
24
On which see S. Strange, Porphyry: On Aristotle Categories (London, 1992).
25
On which see A. Smith, (ed.), Porphyrii philosophica fragmenta (Leipzig 1993)
and now the unpublished dissertation of M. Chase, Études sur le commentaire de
Porphyre sur les Catégories d’Aristote adressé à Gédalios (2000). Stephen Menn
(forthcoming) argues attractively that Gedalius was a Jewish contemporary of
Porphyry.
 Boethus of Sidon 187
III. ON THE Π OF THE CATEGORIES
The Categories is about Simple, Significant
Terms (and Concepts)
Porphyry, following the second-century Peripatetic Herminus,26
reports and endorses Boethus’ view of the skopos of the Categories.
(1) ‘Boethus said what we say’, Porphyry explains: the treatise is about
‘simple significant vocalizations, insofar as they signify realities’ ( æd
çøH ÅÆ
ØŒH ± ºH, ŒÆŁe ÅÆ
ØŒÆ NØ
H æƪ
ø,
Porph. in Cat. 58,4–7). Simplicius also adds Boethus’ view (via
Porphyry) that the Categories concerns words just insofar as they
signify beings (11,23–9). Thus, as Boethus is reported almost
verbatim:
[W]ith regard to nouns and verbs, the division [of the Categories] takes
place as far as the elements of speech (ºª), but according to the
categories the division takes place in so far as expressions (º Ø) have a
relation (åØ) to beings, since they are significant of the latter. ‘This,’
he says, ‘is the reason why conjunctions ( Ø), although they are
to be found within the vocabulary (ºØ), fall outside of the categories.
For they do not indicate any being, not substance, nor the qualified, nor
anything of the kind.’27
This strongly resembles a response to the objections raised by Athe-
nodorus (ch. 5). But Simplicius later adds a slightly different point:
Boethus held that the Categories concerned, not only vocalizations
and things, but the concepts that mediate between them (Simplic. in
Cat. 13,11–18). At 41,28, we also encounter Boethus citing with
approval the view of the ‘ancients’ that only intellections (  Ø)
are signified.28 (Simplicius feels obliged to reply on behalf of the
Neoplatonist consensus in favour of significant expressions.)

26
On whose treatment of the Categories see for example Griffin 2009: ch. 7.
27
æ
ŁÅØ b ŒÆd
a
F B Łı › —æçæØ ººB IªåØ Æ ª
Æ ŒÆd N

e ÆP
e

Æ
E NæÅØ. ºª Ø ªaæ ŒÆd KŒ E, ‹
Ø ŒÆ
a b
e ZÆ ŒÆd
e
ÞBÆ  ØÆ æ Ø N
a
F ºªı ª 
ÆØ 
Øå EÆ, (25) ŒÆ
a b
a ή
Ūæ Æ 
ØÆ æ Ø ª 
ÆØ, ŒÆŁe åØ åıØ ƃ º Ø æe
a Z
Æ, ÅÆ
ØŒÆd

ø
sÆØ. ‘ Ø, çÅ , ƒ  Ø K b
B fi º Ø æ Œ
ÆØ,
H b ή
ŪæØH
KŒ
ıØ· P b ªaæ
H Z
ø źFØ, h
P Æ h
Øe h
¼ºº
Ø

ØF
’ (ed. Kalbfleisch 1907).
28
Boethus also maintained (Simplic. 41,28) that ‘among the Ancients, the only
things said or signified were intellections (  Ø), for truth and falsehood are not in
188 On the Π of the Categories
What generates this slightly different tradition about the reading
that Boethus favours? Porphyry (in his short commentary) and
Simplicius may be preserving a single view in a simplified version
(for beginners), which omits mediating concepts, and a more accurate
version (for advanced readers), which includes concepts.29 I have
tentatively argued elsewhere (Griffin 2012a; 2009: ch. 1) that the
two versions of the Œ —words-things and word-concepts-
things—derive from different original sources, Andronicus and
Boethus respectively. Andronicus may have interpreted the semantics
of the Categories based on Stoic or Epicurean theory, and this inter-
pretation might be largely preserved in the arguments for the title
ή
Ūæ Æ preserved in Porphyry’s shorter commentary (see ch. 2).
In any case, Andronicus’ interpretation did not provide an important
role for mediating concepts. But the semantics sketched in Aristotle’s
De Interpretatione 1 came to dominate in the later tradition, and this
may have started from Boethus. Boethus—who was generally sympa-
thetic to the works of Aristotle as transmitted30—rejected Androni-
cus’ athetization (rejection of Aristotle’s authorship) of the De
Interpretatione, and welded a three-step (words-concepts-things) se-
mantical theory onto the earlier theory (omitting concepts) that
Andronicus favoured. In doing so, Boethus may have inaugurated
the influential ‘conceptualist’ semantical interpretation of the Cat-
egories that Simplicius (and presumably Porphyry in his Ad Geda-
lium) favoured, generating the kind of critique we encounter at
Simplic. 41,28.
Boethus also said that realities are necessarily described by non-
compound expressions (Simplicius in Cat. 41,14–19). Simplicius is
here commenting on a lemma from Cat. 1a16: ‘Of things which are
said, some are said in combination . . . ’ Here, he asks what º ª Æ
might mean. He offers a fourfold answer:31 ‘things said’ might be
(1) realities ( æªÆ
Æ), or (2) notions ( Æ
Æ), or (3) the significant

the realities, but in thoughts ( ØØÆØ) and the developments ( Ø Ø) of the
intellect.’
29
For this interpretation, see for instance Ebbesen 1990, and the masterful treat-
ment of Simplicius in Cat. 12,13–13,11 in Hoffmann 1987.
30
Boethus generally defends against aporiai or enstaseis raised against the text of
Aristotle and the positions adopted by Aristotle, so that it is an exception, rather than
the rule, when he suggests that the text may be corrupt (Simplic. in Cat. 58,23–7).
31
Compare Sosigenes ap. Dexippum in Cat. 7,1 and following, and Moraux 1973:
151–2.
 Boethus of Sidon 189
expression itself, or (4) any expression at all, even a meaningless one.
We might suppose that it was option (3) that Porphyry and Boethus
agreed to be the Π of the Categories.32 At this point, Simplicius
himself does not suggest that we should prefer one of these options,
but cites Boethus on how they might be reduced:
‘What is said’ must be understood in [these] four ways in the case of
non-compound expressions, but in only three ways [sc. thoughts, sig-
nificant expressions, and non-significant expressions] in the case of
those which have been combined. According to him, combined realities,
such as ‘It is day’, are not what is said, not because the ºª does not
signify anything, but because it is not the name of a reality ( æAªÆ), as
day is. It is not of the reality, but about ( æ ) the reality.33 (41,14–19, tr.
Chase 2003, lightly adapted)
Thus the Categories concerns simple, significant expressions qua
significant—but not propositions expressive of states of affairs.

The Categories is about Sensible Beings

(the Referents of Simple, Significant Terms)
Boethus also states that the Categories is not concerned with intelli-
gible being (c ªaæ r ÆØ æd
B Å
B P Æ
e ºª, Simplicius
78,4–5), but with sensible being alone. (I argue, ch. 4, that Boethus is
here replying to Lucius, who is chronologically the earliest figure
named in association with this problem at 73,15–28.) Boethus does
acknowledge the reality of intelligible beings, such as intelligible
number (Simplic. in Cat. 65,20–1).34 Based on his observation that

32
cf. Chase (2003), n. ad loc.
33
Iºº’ ‹ ª BÅŁ K d b
H Iı ºŒø

æÆåH, <‰> YæÅ
ÆØ,
e º ª 
IŒ Ø IØE, K d b
H ή
a ı ºŒc
æØåH · P ªaæ r ÆØ
a
ı º ªÆ æªÆ
Æ, x 
e ‘æÆ K
’ º ª , På ‹
Ø P b ÅÆ  Ø ›
ºª, Iºº’ ‹
Ø PŒ 
Ø
F æªÆ
 ZÆ, u æ
e æÆ. Iºº’ PŒ 
Ø b
F
æªÆ
, æd b
F æªÆ
 (ed. Kalbfleisch 1907).
34
Boethus’ resort to the notion of an intelligible, ‘asomatic’ number has struck
some commentators as surprisingly Platonizing. It certainly shows his direct response,
I take it, to the criticisms of the Categories from a ‘Platonizing’ or ‘Pythagoreanizing’
perspective. Thus Moraux (155) suggests that Boethus is here influenced by Speusip-
pus and ‘an die substantielle Realität von intelligiblen Zahlen glaubte’. Tarán (1981)
argues to the contrary, pointing out that the distinction of two kinds of monads may
be Simplicius’ own (‘then, as Boethus would say [› BÅŁ ¼ çÆ Å], the monad will
also be twofold . . .  ). Michael Chase (2003, ad loc.) proposes that the influence may
go back further still, as Alexander of Aphrodisias (in Met. 56,20) cites a similar
190 On the Π of the Categories
the treatise concerns sensible being alone—a view which could rea-
sonably be derived from the examples of substances (P ÆØ) used in
the Categories35—Boethus suggests that some objections of Lucius
and Nicostratus, and Plotinus (such as whether the Categories amply
deals with the intelligible world) are redundant. This helps to illus-
trate how Boethus responded to criticisms of the Categories as an
account of P Æ.36
A more interesting question, he proposes, involves the relationship
of the accounts of P Æ given in the Categories and the Metaphysics
(78,5–20). This inquiry leads to Boethus’ famous and much-repeated
suggestion (as Porphyry interprets him, at any rate) that form might
qualify as a non-substantial accident of its matter, although in the
context, we might wonder whether Boethus distinguished (perhaps
like Alexander, DA 90,2–15) between sensible form, which can be
understood in this way, and intelligible form. Boethus appears to
acknowledge that entities such as ‘intelligible number’ exist as
P ÆØ—or at least Simplicius believes that he does (65,20–1)—and
can be distinguished from quantities or relations which have the same
name:
‘But what about the one’, they [sc. critics of the Categories following
Lucius] say, ‘and the monad and the point? How can they not fall
outside the categories?’ [ . . . ] If . . . number is twofold—one incorporeal,
the other corporeal—then, as Boethus would say, the monad will also be
twofold: one which is substance (P Æ), and is in intelligible number—
Aristotle also thinks that this one exists—and one which is a relative or a
quantified. Later, however, Boethus says that perhaps it is better to call it
a quantified, for as whiteness is to white, so is the dyad to two. If,
therefore, the former are both qualified, then the latter are also quan-
tified. (Simplic. in Cat. 65,13–24, tr. Chase 2003)37

discussion from Aristotle On the Good (fr. 2 Ross) and references Plato. But in itself,
Boethus’ description of a non-bodily, intelligible nature is not uncharacteristic: at
78,4, Boethus similarly distinguishes between ‘somatic and asomatic’ ousia, and this
passage is analysed further below.
35
Examples like individual human beings and animals would seem to support this
assumption, although examples of non-substances categories—like ‘knowledge in the
soul’—are more difficult to reconcile.
36
For a discussion of Alexander of Aphrodisias’ response, see Ellis (1994).
37
Iººa
e , çÆ , ŒÆd  a ŒÆd
e Å E H PŒ ø
Ø
H
ή
ŪæØH; h
ªaæ  K
Ø, ‰ ¼
ø fi  Ø · h
ªaæ ı å K
Ø, –

I æb Z, h
ØøæØ· A b e j ı åb j ØøæØ r ÆØ æ Œ Ø, ŒÆd
N ØøæØ, j æØ

e j ¼æ
Ø. j ŒÆd ‰ Iæåc
H IæØŁH ŒÆd ‰ 
æ K
fiH
 Boethus of Sidon 191
Boethus is here replying to critics of the Categories, likely Lucius,
who argue that the number one and the geometrical point have
no place among the ten categories described by Aristotle—which
looks like a version of their standard objection that the Cat. has
ignored intelligible P Æ. Boethus’ response might mean some-
thing like the following. Suppose I use the word ‘one’ to pick
something out. I might pick out a substance: in that case, I am
presumably naming the monad, something intelligible. Alterna-
tively, I might use ‘one’ to pick out a sensible quantity, if I mean to
refer to the quantitative singularity of something here in front of
me (say, Socrates). In either case, I am referring to something
that falls under the list of ten categories. But Boethus maintains
that strictly speaking (cf. 78,4–5), the Categories is not about
intelligible items, but about sensible items. Therefore, ‘perhaps it
is better’ (as Boethus continues here) to locate ‘one’ in the category
of quantity.
There are a number of puzzles about this passage: for example,
where does Aristotle maintain that an intelligible, substantial number
exists? Simplicius’ editor, Kalbfleisch, suggests ad loc. that the refer-
ence may be to An. Post. 1.27, 87a26, but as Chase 2003: 147 n. 735
suggests, the reference may be to an earlier Pythagorean position
described by Aristotle in On the Good. And again, why should
Boethus endorse what looks like a Platonist view by entertaining the
existence of intelligible, substantial number? Moraux (1973: 155)
proposes that Speusippus, who seems to have influenced Boethus’
approach to the Cat., may lie behind this (but see Chase 2003: n. 176).
A passage of Dexippus (in Cat. 45,3–30) might also be read as
suggesting that Alexander and Boethus shared the view that the
Categories is not about intelligible beings, but that there nevertheless
are such beings (but against this interpretation, which I endorsed in
Griffin 2009: ch. 5, see now Chiaradonna 2013).38 (I will not go

æ
Ø ŒÆ
Æ
ÆåŁ 
ÆØ, ‰ ŒÆd
fi H º  æø fi Œ E. N b Ø

e › IæØŁ, › b

IÆ
, › b øÆ
،, 
ÆØ, ‰ ŒÆd › BÅŁ i çÆ Å, ŒÆd  a Ø

,  b
P Æ,  K
fiH Å
fi H (20) IæØŁfiH (
F
 b ŒÆd æØ

º Ø Œ E),  b æ
Ø j
. o
æ  çÅØ › BÅŁ  
¼ Ø r ÆØ e çÆØ· ‰ ªaæ º ıŒ
Å
å Ø æe º ıŒ, o
ø ŒÆd  ıa æe
a · N s KŒ EÆ ¼çø Ø, ŒÆd
ÆF
Æ
 (ed. Kalbfleisch 1907).
38
I argued for this interpretation in Griffin 2009: ch. 5; but see Chiaradonna 2013
for a different, careful, and thoughtful reconstruction of the passage.
192 On the Π of the Categories
further into this question here, but for Boethus’ metaphysics more
generally, see now Rashed 2013 and 2007; on Rashed’s view, Boethus’
position is more materialistic than Alexander’s, and Alexander reacts
against him.)39
For reasons of space and focus, I omit here Boethus’ careful
discussions of a range of other themes in the Categories, including
the Relative (on which he wrote a full book, Simplic. 163,6, which is
cited here and there by Simplicius through 159,14–202,1), Acting and
Undergoing (302,16), When and Where vs. Time and Place (348,2,
and see Huby 1981 for the attribution of another fragment on the
When to Boethus, with caution in Rashed 2013: 71), and on Having
(373,8), which I addressed in more detail in Griffin 2009. Still, it is
worth stressing, if only briefly here, that Boethus defended a number
of traditional doctrines of the Categories. He contended against
innovations in the category of the Relative. He maintained the
division of Acting and Undergoing as independent categories. He
maintained the categories Where and When rather than Place and
Time, perhaps, as Huby suggests, in the course of a dialogue with
Andronicus which would later undergo revision at the hands of
Plotinus. He maintained the category of Having, against the Stoics.
On the other hand, he also seems to have credited some innovations
as beneficial, or at least recognized developments outside the
bounds of the text: for instance, he suggested that Speusippus, as
well as more recent Stoics (
E  ø
æØ), had developed a
more complete collection of ‘onymies’40 than that represented at
Cat. 1a1–15.

39
With respect to Alexander’s view on universals, Tweedale (1984) suggests that
Simplicius—and other modern commentators who deny that Alexander ever treated
universals as ‘primary’—have simply misunderstood Alexander, and perhaps based
their interpretations upon an insufficiently nuanced reading of the primary texts.
Following Paul Moraux (1942) in some points, and himself followed in the main by
Dillon (1990) and Sharples (2005), Tweedale takes S. Pines and the later Neoplatonic
tradition on Alexander’s universalism to task as follows (1984:290–1).
40
That is, linguistic phenomena like homonymy, synonymy, and paronymy,
which classify objects according to criteria such as whether they share a name or a
definition. Boethus also took an interest in classifying the many meanings of words
such as ‘genus’, ‘species’, and ‘differentia’, much as Porphyry did in the Isagoge
(1.18–2.10, etc.): see Chiaradonna, Rashed, Sedley, and Tchernetska 2013, 14,4 and
commentary p. 184.
 Boethus of Sidon 193
IV. THE STRUCTURE OF BOETHUS’ WORK:
I æ ÆØ AND º Ø
The clearest instances41 of I æ ÆØ to which Boethus responds by
name can be credited to Lucius, or rather to the ‘followers of Lucius’
(ƒ æd
e ¸ŒØ). As we noted above, Lucius is mentioned by
name on eight occasions in Simplicius. In each case, Boethus is
subsequently cited to deliver a reply. Porphyry is generally on the
scene, frequently to boost the view propounded by Boethus, but
occasionally to deliver an additional critique. Leaving aside the
ambiguous cases,42 some solid examples of Lucius’ I æ ÆØ in Sim-
plicius are as follows (for what follows see also ch. 4):43
(1) 48,1. The followers of Lucius question how what is in a subject
can be so ‘not as a part’. After a solution from Porphyry, and a
comment by Simplicius (49,1), the I æ Æ continues, culmin-
ating in the absurd conclusion (49,31) that Socrates appears,
on Aristotle’s argument, to be an ‘accident of place’. Finally,
Boethus is cited to offer his own solution to the dilemma
(quoted in ch. 4).
(2) 64,18. The section is headlined by the observation that Cornu-
tus, Lucius, and Nicostratus all objected to the division of the
ten categories (62,27). The best manner of response to these
critics is ‘by question and answer’. ‘The followers of Lucius’
question why Aristotle omitted conjunctions from the Cat-
egories ‘if they are also significant expressions’. Simplicius
delivers a reply, to which Lucius again furnishes a rejoinder:
‘What about the one, the monad, and the point?’ Boethus,
finally, replies that the monad is twofold.

41
An interesting example of his modus operandi is Simplic. in Cat. 348,2–7, a
passage that Huby reads as a submerged dialogue between Boethus and Andronicus: if
this is correct, it is similar in structure and style to Porphyry’s treatment by Question
and Answer, in that it does not appear to represent a one-way criticism of a
predecessor’s views, but a dialectic debate.
42
There are also unnamed instances, such as 58,27, where Boethus ‘gives way’ to
an aporia (K f
B fi I æ fi Æ, cf. 58,23–7) which may also have been posed by the
school of Lucius.
43
Of the eight citations of Lucius ascribed to Simplicius, this list of five omits 1,19
(where he is merely mentioned as the predecessor of Nicostratus) and collapses 62,27
and 64,18 into (2).
194 The Structure of Boethus’ Work
(3) 73,28. The followers of Lucius, in the company of Nicostratus
and Plotinus (6.1), question whether ‘Aristotle here divided
and enumerated sensible and generated things alone, or
whether he did so with all beings . . . and in general whether
the intelligible genera are different from the sensible’. Sim-
plicius replies that ‘he is primarily discussing things of
this world . . . which are those immediately signified by
words’ and observes that we should focus on the ‘threefold
division’ of form, matter, and the compound (347,19–348,2);
a source for this reply is not presently forthcoming, but the
same view recurs within a few pages under the name of
Boethus (348,2).
(4) 125,16. The followers of Lucius criticize Aristotle on the
grounds that he transfers body, which belongs to substance,
to the category of quantity. On this occasion, the reply pre-
cedes the citation: Lucius’ school is ‘mistaken’ to make this
criticism, because ‘a body, insofar as it is extended three ways
and can by nature be measured, is a quantity; but insofar as it is
a substrate, remains the same and one in number and is
receptive of the contraries, in that respect it is a substance.’
An allusion to a response also found in Porphyry’s commen-
tary (in Cat. 103,18 ff.) follows. But where does this reply come
from originally? Each of the doctrines is maintained by
Boethus elsewhere in a manner which suggests a reply to this
I æ Æ: (1) an enmattered quantity can be regarded in two
different ways, as substance or as quantity (65,20); (2) it is a
substance in so far as it is receptive of the contraries and one
and the same in number; (3) it is measurable.
(5) 156,17. The followers of Lucius contend that quality is nearer
(more NΠE) to ousia than is the relative, observing that it is
better to identify Socrates by his qualities than by his relations.
The priority of quality holds, they continue, if one divides the
categories into ŒÆŁ Æ
 and æe 
æ. To this Porphyry
replies that the relative should come next, arguing (in Cat.
111,9–112,5) that (1) when length, depth, and breadth have
come into existence, the larger and smaller supervene upon
them, and these are relatives; (2) also because Aristotle needs
to explain what relatives are, having introduced them in
discussing quantity. Porphyry continues to explain what
 Boethus of Sidon 195
relatives are, concluding with an observation, namely, that the
definition given by Aristotle here is Platonic and was subse-
quently corrected. Simplicius (159,12–15) ascribed this correc-
tion to Boethus, which may imply that Boethus underlies
much of this discourse, which is presented more fully by
Simplicius at 157,23–158,27, also immediately preceding a
citation of Boethus (159,14).
The pattern is consistent, and may be broadly sketched as follows
here (but see the ‘table of responses’ in ch. 4 for an overview): (a) ‘the
followers of Lucius’, sometimes followed by Nicostratus or Plotinus,
present a problem with the text of the Categories, which is taken up in
sequence with the ή
a ºØ exegesis; (b) Porphyry provides a
response; (c) Simplicius provides a response, which may be his own,
but sometimes appears to be a submerged source; (d) Boethus pro-
vides a response, with Simplicius typically citing him by name.
Elements (b) and (c) are sometimes reversed. I think it is highly
probable, in light of the considerations mentioned earlier, that
element (c) typically represents the opinion of Iamblichus in his
own commentary On the Categories, drawing in places on the text
of ‘Archytas’. This pattern integrates and develops a ‘back and
forth’ dialectic that sometimes resembles, albeit with greater matur-
ity, the style and content of Porphyry’s commentary By Question
and Answer. The same pattern may also be reflected in the
dialogue ascribed to Andronicus and Boethus by Huby (1981) at
347,19–348,2.
This simplified stratigraphy suggests, in my view, several results.
First, Simplicius’ most immediate sources are Iamblichus’ commen-
tary On the Categories, the source for element (c), and, beneath it,
Porphyry’s commentary Ad Gedalium, as he informs us in the pre-
amble to his commentary. Second, the parallel passages in Porphyry’s
Ad Gedalium contained at least the following elements: (a) the view
taken by the followers of Lucius (sometimes advanced by Nicostratus
or Plotinus), (b) views expressed by ‘Porphyry’ himself, perhaps (as
I suspect)44 encapsulating a history of other predecessors’ works, such

44
The closest referent in Porphyry’s in Cat. answering to the dialogue discussed by
Huby is not, in fact, Boethus, but Herminus, who is often mentioned by Porphyry as
sharing Boethus’ views. I discussed Herminus’ role in the transmission further in the
opening chapter of Griffin 2009.
196 The Structure of Boethus’ Work
as those of Herminus,45 and (d) Boethus, in that order, sometimes
followed by (e) a correction of Boethus’ view by Porphyry himself,
surviving for us in fragments such as Simplicius in Cat. 78,4. It may
be deduced that Boethus often had the last word in Porphyry’s
treatment, at least when Porphyry and Iamblichus were content
with his position, a situation which persisted throughout much of
Simplicius’ treatment.
Why was this structure adopted? So long as the commentators
were sympathetic to the solution maintained by Boethus, the stratig-
raphy in Simplicius suggests that Porphyry’s and Iamblichus’ com-
mentaries successively interpolated content between Lucius’ I æ Æ
and Boethus’ answer, generating elements (b) and (c). They main-
tained the integrity of Boethus’ treatment as a ºØ of the I æ Æ,
and did not amend it, but occasionally added a corrective response
when Boethan IªåØ Æ fell short, so to speak, of the Platonic
truth.46
I infer that Boethus’ commentary on the Categories ŒÆ
a ºØ may
be the original source both for the Lucian I æ ÆØ and for Boethus’
responses to them. This commentary, like Porphyry’s, Iamblichus’,
and Simplicius’ major works, treated the text in sequence, probably by
lemmata, and investigated certain I æ ÆØ in the form of questions
and answers. Many of these apories were traced by Boethus himself to
the Lucian school, which he sought to answer. (Some of them may
also have been raised by Andronicus; cf. Huby 1981). In general,
Boethus’ responses to these puzzles generated the content of elements
(a) and (d) that survive in Simplicius, and perhaps many sections of
unsourced dialogue.
Thus the stratigraphy of the underlying commentaries—each mod-
elled on its predecessor and incorporating some new material—might
follow this model:

45
The teacher of Alexander who followed Boethus’ views on the skopos of the Cat.
for instance, ‘albeit briefly’ (Porph. In Cat. 59,17–18).
46
This is particularly intriguing in the context of Michael Chase’s (2003: 94)
suggestion that Simplicius ‘admired B[oethus] more than any other pre-Porphyrian
exegete’. The epithet thaumasios bestowed upon him is exceptional, and in general
the attitudes of the commentators imply that he was treated with some of the
caution and respect accorded to a genuine Neoplatonist commentator. One might
venture to suggest that Simplicius’ sources’ attitudes differed slightly, as one shies
away from a direct criticism of Boethus’ ‘ignorance’ (149,14) while the other is
prepared to debunk it (163,3 and 163,15); in general, however, his arguments are left
intact.
 Boethus of Sidon 197

Simplicius
Iamblichus (Incorporating: Pseudo-Archytas)
Porphyry (AG) (Incorporating: Nicostratus, Athenodorus, Cornutus, Herminus,
Plotinus, and others)
Boethus (Incorporating: Lucius, Andronicus)

Did Boethus, like Porphyry, compose two commentaries, one dir-

ected to a real or imagined pupil and adopting the form of questions
and answers, the other directed against I æ ÆØ raised by historical
objectors and adopting the ή
a ºØ form of close exegesis? It may
be equally likely that these two formats were used in a single work.
Porphyry’s Question and Answer commentary, once it gets underway,
develops its comments based on the quotation of individual lemmata,
most of which are cited by his interlocutor (see for example 86,33–4),
and many of which cite named historical figures such as Atheno-
dorus, Cornutus, and Herminus. Conversely, Simplicius’ pattern of
presentation demonstrates that a ή
a ºØ commentary can also
incorporate a dialogue format presenting aporetic questions and
answers; indeed, Simplicius or his source, at 63,2, recommend the
genre of ‘debate’ (
e
 
B I
غª Æ) to address the I æ ÆØ
raised by Lucius and his successors.
Porphyry’s commentary By Question and Answer is primarily
pedagogical, and represents a work simpler than, but analogous to,
his monumental treatment Ad Gedalium.47 Both commentaries
address interlocutors who are otherwise unknown (but on Gedalius
see now Menn forthcoming). Lucius is also a figure otherwise
unknown: I speculated earlier (in my discussion of Lucius and Nicos-
tratus) that he may be an interlocutor comparable to Porphyry’s
student in the Categories commentary.48 Such speculation could not
easily be supported against the testimony of the later sources, and
particularly Simplicius, who views Lucius as an historical figure and
the predecessor of Nicostratus.

47
This could be argued citing comparanda in Porphyry’s commentary by question
and answer under the headings, for instance, 90,29–81,12, 95,16–38, 111,5–112,7;
I argued along these lines in Griffin 2009: ch. 5.
48
But given the poverty of the evidence, such speculation treads into the treach-
erous territory of replacing Homer with ‘another man of the same name’!
198 Conclusions
V. CONCLUSIONS

The stratigraphy discussed above suggests that Boethus’ responses to

aporiai, including Lucius’ problems and (perhaps) certain novel inter-
pretations proposed by Andronicus, represent a general approach to
the exegesis of the text that would go on to exercise considerable
influence on the structure and content of future treatments. However,
our review also suggests that Boethus’ solutions were not immediately
adopted; in fact, the shadowy commentators following Boethus, such
as Nicostratus, appear to have challenged his answers to Lucius, and
adopted and developed these aporiai as increasingly serious chal-
lenges. These succeeding philosophers who adopted and pursued the
positions ascribed to Lucius may, perhaps, have been broadly charac-
terized as ƒ æd
e ¸ŒØ, the group to whom Simplicius ascribes
each of the relevant aporiai cited above in the Categories commentary.
In the chapter on Lucius and Nicostratus, we also considered this
tradition of commentary that immediately followed Boethus. The
tradition appears to have been defined by engagement with Lucius’
aporiai as well as Boethus’ exegesis and solutions, and these responses
evolved over time into a positive redefinition of the problem con-
cerning the genē of being, represented in the culminating point of
Plotinus’ treatise 6.1–3.
We do not often find Boethus engaging directly or frequently with
the ‘linguistic’ view of the Categories presented by the Stoic Atheno-
dorus, except for the occasional but telling remark that the text
doesn’t address conjunctions because its target is semantics (Simplic.
11,23–9). However, as Porphyry’s shorter commentary shows, Her-
minus did treat Boethus as maintaining a view capable of withstand-
ing Athenodorus’ objections, and Herminus himself sought to rebut
those objections.49
Boethus himself defended a view of the Categories according to
which it had something to say about the structure of being itself, and
not merely about language. His most explicit polemical concern is to
deal with the objections of the ‘Platonizing’ critics such as Lucius. Later,
Herminus would adopt the formula that the Categories concerns
‘simple words significant of beings, insofar as they are significant’,

49
See also Griffin 2009: ch. 1.
 Boethus of Sidon 199
and ascribe this formula to Boethus, but deploy it primarily in defence
against the ‘linguistic’ criticism promoted by Athenodorus.
This chapter has very briefly surveyed the much larger plain of
Boethus’ contribution to the tradition of philosophical commentary
on the Categories.50 Boethus does emerge as an historian of philoso-
phy and as an original thinker who provoked strong responses among
his contemporaries and in the school of later Neoplatonism that
followed Porphyry. He developed the thought of certain predecessors,
including Andronicus, in certain areas, but his greatest contribution
was to maintain and revitalize a strong line of Peripatetic thought in
an idiom and an intellectual context which was clearly understood by
Stoics and Platonists as relevant, and inspired a re-engagement,
sometimes contentious and sometimes familiar, with the esoteric
treatises of Aristotle (discussed in ch. 2). In short, Boethus trans-
formed the Categories into a rich repository of content that would
occupy centuries of discussion. He raised critical challenges of inter-
pretation that remain with us today, such as the problem of recon-
ciling the Categories’ notion of ousia with the hylomorphism of the
Metaphysics. Lastly, Boethus asked why Aristotle subdivided ousia
into three kinds in the Metaphysics, namely Form, Matter, and the
Compound, yet circumscribed ousia in a single genus in the Categor-
ies. This question is one of Boethus’ most lasting, and challenging,
contributions to the tradition.

50
I attempted to treat that contribution in more detail in Griffin 2009: ch. 5; see
now Rashed 2013 and 2007 for outstanding introductions to Boethus’ metaphysics,
and Chiaradonna 2013 for a reinterpretation of a crucial text in Dexippus.
 7

The Second Century

A Brief Overview1

This chapter briefly surveys subsequent developments in the tradition

on the Categories.

I. SECOND-CENTURY PLATONISTS

Already following Plutarch (discussed briefly in ch. 3), by the middle

of the second century ce, we find in Alcinous’ Didaskalikos examples
of Platonists using the ten categories. The author claims that Plato
‘gave indications’ of the ten categories in the Parmenides (although as
Dillon 1993: 84 points out, the verb
Ø suggests that Plato is
less than explicit).2 We can see here a continuation of the spirit (if not
the letter) of Eudorus’ warmth to the Categories in principle, and
Antiochus’ ecumenical interest in the common dialectical tools of the
Old Academy.
On the other hand, Atticus, who was active under Marcus Aurelius,
takes a decidedly polemical line towards Aristotelianism. He wrote a
critical tract ‘Against those who interpret Plato’s philosophy by
means of Aristotle’, some of which is preserved by Eusebius (Praep.

1
A fuller list of the contemporary personae may be found in Sorabji and Sharples
(2007).
2
Proclus in Parm. 1083,37 would also later attribute to certain ‘earlier’ commen-
tators the view that, in the first hypothesis of the Parmenides, the ten categories were
denied of the One; see Did. 6.10, and Dillon 1977: 279 and 1993: 84–5 for commentary
and list of relevant citations in the Parmenides itself.
 The Second Century 201
Ev. bks 11 and 15).3 With respect to the Categories, Atticus is said by
Simplicius (in Cat. 30,16; 32,19 = frr. 41–2 Des Places) to have offered
two comments on the definition of homonyms and synonyms at the
outset of the Categories. In the first, he elaborates the objection of
Nicostratus; the second focuses on the relationship of analogical and
metaphorical uses of terms. Atticus represents an approach different
to that of, say, the author of the Didaskalikos, who seeks to show that
Plato is the source of what is useful in Aristotle. In taking this line,
Atticus appears to follow L. Calvinius Taurus (second quarter of the
second century), who wrote ‘on the doctrinal differences’ of Plato and
Aristotle (Suda T 166, v. iv p. 509.12 Adler).4

II. PERIPATETICS

In the early second century, Aspasius—who taught Herminus, who in

turn lectured to Alexander of Aphrodisias—authored a commentary
on the Categories; he (among many others) may have mediated the
Boethan commentary tradition to Herminus and Alexander, import-
ant sources for Porphyry.5
We also know from Galen (De libr. propr. 42,20–43,1) that Adrastus
of Aphrodisias wrote a commentary on Categories in the second
century ce.6 Adrastus appears to have re-instituted the attachment
of the Categories to the Topics (and presumably re-emphasized its
value for dialectic in the Topics sense), while retaining its Andronican
place at the outset of the curriculum. His arguments were evidently
contained in a treatise entitled ‘Concerning the Order of Aristotle's
Treatises’ (Simplicius, in Cat. 16,1–4). Adrastus also revived the title
‘Before the Topics’ for the Categories—this may suggest that he was

3
Broadly, he argues against Aristotle’s doctrine of the telos, his alleged disbelief in
pronoia, his teaching on the eternity of physical world, on the fifth ousia and motion
of the heavens, and his supposed denial of the immortality of the human soul, in
addition to Aristotle’s denial of the Platonic world-soul and eidē.
4
See Dillon (1977: 237 ff.), Dorrie and Baltes, 310–23, and Praechter 101 ff.
5
Indeed, Aspasius authored a series of commentaries, on the Categories, De Int.,
Physics, De Caelo, Metaphysics, and Nicomachean Ethics. Of these, just the last is (in
significant portions) extant. For our purposes, he is a key source for the views of
Andronicus and Boethus on ÆŁ .
6
And from Simplicius (in Phys. 122,33) that he wrote a commentary on Physics.
202 Peripatetics
responding directly to Andronicus (cf. Simplic. 15,30 and 16,14).
Adrastus also discussed a second ‘version’ of the text of the Categor-
ies,7 suggesting that he had access to a diversity of textual traditions.

Alexander of Aphrodisias, His Teachers

and Contemporaries
Simplicius, early in his commentary (13,11–18), offers this ‘com-
promise’ view about the Œ  of the Categories, which is attributed
to Alexander of Aphrodisias as well as several of his teachers, sug-
gesting the development of a consensus in the later second century:
It is thus clear from the preceding considerations that the goal [of the
Categories], appropriate to the study of logic, is about simple, primary,
and generic words (çøÆ), in so far as they are significant of beings.
Instruction is, at any rate, also given about the realities which are
signified by them, and about notions ( ÆÆ), in so far as it is realities
that are signified by words. This is also the view of the Alexanders
[i.e. Alexander of Aphrodisias and Alexander of Aigai], Herminus,
Boethus, and Porphyry; it was seconded by the divine Iamblichus,
clarified by Syrianus, and accepted by our teachers. (tr. Chase 2003)8
Alexander himself is the central figure in this list, certainly the most
prominent before Porphyry. Simplicius’ framing of the consensus
suggests (chronologically) that Alexander’s treatment of the Categor-
ies integrated the views of Alexander of Aigai and Herminus, and
Boethus (perhaps representing a commentary tradition with multiple
layers dating back to Boethus: cf. Griffin 2009: ch. 1). Here we might
say a few words about Alexander’s teachers, the link between the
earlier commentary tradition that we have been treating, and
Alexander. (Alexander himself falls outside the scope of this study;
but on his role as a source, see also Appendix 1.)

7
On him, cf. Simplic. in Phys. 4.11, 6.4 ff; in Cat. 15.36, 18.16 ff.; cf. Zeller III 1, 809
n. 3; Moraux Listes 58 ff.
8
 Ø s KŒ H NæÅø Bº , ‹Ø K d b › Œ e NŒE B fi º ªØŒB fi
æƪÆÆfi æd H ±ºH ŒÆd æø ŒÆd ªØŒH çøH, ŒÆŁe ÅÆØŒÆd H
Zø N , ıØ ŒÆØ b ø ŒÆd a ÅÆØ Æ
’ ÆPH æªÆÆ ŒÆd a
 ÆÆ, ŒÆŁe ÅÆÆØ a æªÆÆ
e H çøH. ŒÆd ÆFÆ ŒÆd  E
ºæ Ø ŒÆd  Eæø fi ŒÆd B Łø fi ŒÆd — æçıæø fi  ŒE ŒÆd › ŁE  ºØå
KØłÅçÇØ ŒÆd ıæØÆe ÆçÅÇØ ŒÆd ƒ æ Ø Ø ŒÆº Ø I å ÆØ (ed.
Kalbfleisch 1907).
 The Second Century 203
Herminus, who lectured to Alexander sometime in the later second
century ce, is a particularly important figure in the transmission of
our story.9 His Aristotelian lectures were heard by Alexander of
Aphrodisias,10 and he was described by his contemporary Lucian11
as a man who ‘praises Aristotle, with a mouth full of his ten categor-
ies’. As Porphyry’s report of him clearly shows (in Cat. 59,17–33),
Herminus endorsed the view that the Categories was a suitable intro-
duction to philosophy, based on its ability to help articulate our
natural æ ºÅłØ of the ten genera. I have argued elsewhere that
Herminus was important for mediating the earlier tradition to Alex-
ander, through whom Porphyry has his own reports of Herminus’
views.12 I have also speculated that Herminus adopted a more ‘An-
dronican’ version of the Œ  of the Categories, as addressing words
that directly signify things (without conceptual mediation).13
Also in the later second century, Sosigenes14 lectured to Alexander
(Alex. in Meteor. 143,13: › Ø ŒÆº H ø تÅ ).15 Sosigenes
framed the debate about the Π of the Categories in a way that
might have been particularly influential in the later commentary
tradition, asking what Aristotle means by ºª  , ‘what is said’:
‘word’ (çø ), ‘thought’ ( ÅÆ), or ‘object’ (æAªÆ) (cf. Simplic. in
Cat. 7,4–9,24; Dexippus 7,4). Sosigenes himself did not choose

9
On Herminus see Gottschalk (1987), 1158–60 and (1990), 79; Moraux (1984), II,
361–98; and Goulet (2000), III, 652–4. His fragments were collected by Schmidt
(1907). Modern students of Aristotle’s logic have noted and criticized his views on
the Prior Analytics and the Topics; see for example Łukasiewicz (1951), 30–1; Patzig
(1968), 118–22; Smith (1997), 141–2; Barnes et al. (1991); Flannery (1994); and
Bochenski (1947), 64. For earlier studies on Herminus, see Prantl (1855–70), I,
545–6 (Categories), 584–50, 552–3; 555–7 and Zeller (III 1, 778, 1 and 783, 2).
10
Simplicius, De caelo 430,32–431,11 reports a comment of Alexander that
 Eæ ı  XŒ ı Æ, ŒÆŁa q ŒÆd K  E  Æ  ı çæ  . Simplicius adds ÆFÆ
b s a  F  Eæ ı ŒÆa e   Ø  NæÅÆ. See also Moraux (1984), II,
361–3. As I suggest below, Herminus may have been known to the later commenta-
tors solely, or chiefly, through Alexander.
11
Life of Demonax 56,1–6: e æØ  ºÅ ’ KÆØ FÆ ŒÆd Øa  Æ ÆP F
a ŒÆ ŒÆŪ æÆ å Æ. Lucian appears to imply that his readership would, like
Demonax (Ng ªaæ . . . , 56,2), already be familiar with this Fachgebiet of Herminus.
12
Griffin 2009: ch. 7.
13
Griffin 2009: ch. 1, 7; Griffin 2012a.
14
On him see Zeller 1990 (orig. 1856): 813; Rehm 1927, and Moraux 1973.
15
He wrote on optics, as Alexander tells us, and he wrote a detailed account of the
theory of Eudoxus and Aristotle, and Callippus about the planetary spheres, explain-
ing the use of epicycles (see Simplic. in Cael. 493–510; also Alex. in Met. 703–6).
204 Peripatetics
between the three theses,16 but his scholastic way of carving up the
argument fits the doxographical tendency of the later commentators.
Coming to Alexander’s prodigious contemporary Galen
(129–c.216 ce), we might just note in passing a few portions of his
output of direct relevance to the Categories. First, Galen tells us of his
own commentary, which does not survive, but demonstrates an
interesting attitude towards his predecessors—an attitude of respect,
which urges the reader to privilege the earlier commentaries on the
treatise. We also meet this attitude in Simplicius when he enjoins us
to prefer the commentaries of Iamblichus and Porphyry to his own.
Galen writes:
On [Aristotle’s] work on The Ten Categories I had not previously
written any commentary, either for myself or for others; when subse-
quently somebody asked me for something on the solution posed on
that work, <I wrote a commentary> with the firm instruction that he
should only show it to students who had already read the book with a
teacher, or at least made a start with some other commentaries, such as
those of Adrastus and Aspasius.17 (On My Own Books 19.42–3 Kühn,
translation after Singer)
Second, some fragments of Galen’s own views on the categories do
survive. In the first place, we know that he retained the ‘ten’ categor-
ies, but also added an eleventh, ‘synthesis’, which he suggests Aristotle
omitted in error (Inst. Log. 13); thus, from Simplicius’ standpoint,
Galen would stand among those who, like Athenodorus, judged the
tenfold division of the categories deficient.18 But this all goes to show
that it was natural for a second-century polymath like Galen to
comment on the Categories and to expect that they should be read
by relative philosophical beginners with the help of a commentary
and a teacher.

16
cf. Moraux in Hermes 95 (1967), p. 169 on Alexander.
17
 F b H ŒÆ ŒÆŪ æØH PŒ K ØÅ Å h’ KÆıfi H Ø  Ø F 
 ÅÆ
æ Ł hŁ’ æ Ø øŒÆ ŒÆd Øa  FŁ’ o æ    H Ææø Ød ÅŁØ

  Æ’ (15) åØ <æØå Ł’>, ‹ Æ [ŒÆa e غ ] <I> Œ ı ØN a H K
ÆPfiH ÇÅ ıø º Ø ***, Œ Øø E KŒºı Æ H
 Åø KŒ Ø   Ø
 E IªøŒ Ø Ææa ØÆ Œºø fi e غ  j ø ª æ Ø Åª Ø Ø’ æø
KŪÅØŒH, › EÆ  ’ æ  ı ŒÆd  Æ  ı K  (ed. Helmreich et al. 1891).
18
Elias (in Cat. 160,20) also ascribes a list of five categories (ousia, poion, poson,
pros ti, pros ti pōs echon) to Galen, although Prantl (I 564 n. 85) takes this to be an
error.
 8

Conclusions

Our Categories was perhaps not unknown at the dawn of the first
century, although it would have been known under a different title,
—æe H ø, and interpreted perhaps as a source of a familiar,
handbook list of basic question-types (similar to Quintilian’s inter-
pretation of the ten elementa at Inst. 3.23–34), or as an aid for
correcting sophistical arguments based on category confusion and
developing tests for finding the genus of a term (with criticism offered
by those who troubled to interpret the text in detail at all, perhaps
including Stoic philosophers like Athenodorus, who read it as a
division of basically linguistic items).
As Aristotle’s esoteric works began to attract some new interest in
the limited circles represented by men like Tyrannio, and among
Peripatetics like Andronicus and Ariston, we might see Andronicus
as especially struck by the value of the Categories as a proleptic
treatise able to ‘tease out’ and articulate our preconceptions. The
Categories triggered this process at the level of ordinary language,
illuminating the simple kinds relevant to definition and division. We
might further speculate, based on his interest in recognizing and
correcting equivocation in the Categories, that Andronicus was also
interested in its utility for resisting sophistical arguments that threat-
ened the possibility of dialectic and demonstration.
Andronicus, then, brought the treatise to the attention of a wider
audience through his catalogue, paraphrase, and other work. Eudorus
exemplifies that interest—representing a growing Academic, ‘Platoniz-
ing’ school of interest in Alexandria that makes the Categories not only
a useful tool but a map of intelligible reality. That view sparked rival
critique from Platonist quarters, including the criticisms offered by
Lucius. Andronicus’ pupil, Boethus of Sidon, replied to these criticisms
(engaging in turn in a more detailed, metaphysical interpretation of the
206 Conclusions
Categories), but also replied to the linguistic reading of the Categories
represented by Athenodorus, of whom Andronicus may not even have
been aware, and at any rate by whose worries he does not seem to have
been much bothered. Boethus’ compromise interpretation, that the
Categories concerns ‘words significant of things, qua significant’—
with the further rider that the ‘things’ initially signified by our ordinary
words are sensible things—responds effectively to both ‘Lucian’ and
‘Athenodoran’ lines of criticism. He explains why the Categories’ scope
is restricted, whether it is treated as a work about language or being,
and he introduces a fundamentally new way of understanding the
Categories as a treatise about semantics. This Boethan reading would
prevail until Porphyry, with, in the intervening years, Nicostratus and
Plotinus taking up Lucius’ problems, and Cornutus also responding to
Athenodorus; Iamblichus would perhaps return, as briefly discussed
below, to the Eudoran view that the Categories might offer a useful
map of intelligible being. (For the difference between Iamblichus’
and Porphyry’s interpretations of the Categories, see Chiaradonna
2007b; I also discuss the interrelationship of the later sources in Griffin
2009: ch. 1.)
This is the story that I have primarily sought to trace in this
book. But interest in the Categories continued to develop. Later, the
Stoic division of the first and second parts of dialectic may have
influenced the second-century Peripatetic Herminus in his influential
distinction of the ‘first’ and ‘second’ imposition of language, itself
constructed around earlier Hellenistic theories of language. The ‘sec-
ond’ imposition, after all, concerns discussion of linguistic expres-
sions in themselves (the second compartment of Stoic dialectic),
while the ‘first’ imposition concerns discussion of the entities that
they signify, insofar as the linguistic expressions signify (the first
compartment of Stoic dialectic). Herminus’ defence against the
Stoic attacks on the Categories, and his revitalization of Boethus’
formulation of the skopos, may represent the inauguration of the
‘story’ told by Porphyry, where some historical figures were partisans
of the Categories as a (poor) account of being, others of the Categories
as a (poor) account of language, and only Boethus and Herminus got
it right as a (good) account of both. Certainly Herminus conveyed,
and perhaps inaugurated, the view that the Categories was a peda-
gogically suitable introduction to the quest for knowledge. But even
as Herminus lectured, Sosigenes discussed an alternative ‘history’,
namely a ‘tripartition’ of historical sources that included a mediating
 Conclusions 207
‘conceptualist’ view of the subject matter of the Categories. Unlike
Herminus, however, Sosigenes did not take a position on which of the
sources was correct.
Alexander of Aphrodisias, I think,1 incorporated both of his lec-
turers’ views—Herminus’ and Sosigenes’—on the Categories. Por-
phyry, then, adopted Alexander’s account wholesale in his own
commentary in Cat., including both Herminus’ defence and Alexan-
der’s critique of it. It is fair to presume that Porphyry added and
subtracted considerable detail and innovation, perhaps partly motiv-
ated by the approach taken toward the Categories, or rather toward
the exegetical tradition on the Categories, in the school of Plotinus. In
the longer commentary Ad Gedalium that served as a basis for future
commentators, there was room for a comprehensive discussion of the
positions reported and taken by Alexander. In the shorter commen-
tary in Cat. that has survived, there was room only to report Hermi-
nus’ position, and (in keeping with Porphyry’s own practice of
simplification, discussed early in the Isagoge) to omit difficulties
about the real nature of the entities discussed in the Categories.
Iamblichus2 then came to add both ‘noeric theōria’ and the views of
‘Archytas’ to Porphyry’s commentary. In so doing, we may note,
Iamblichus was in many ways reverting to an alternative, non-
dialectical reading of the Categories that originated in the first century
bce, at least as early as the Andronican reading, with Eudorus,
‘Archtyas’, and Lucius, and which is preserved in Plutarch (An.
Proc.) and Philo, among others who also viewed Plato as an early
proponent of this use of the ‘ten categories’. But Iamblichus reapplied
this exegetical approach to the history of scholarship on the Categories
as well as to the text itself, describing the ‘tripartition’ of views on the
Œ as significant of three ‘levels’ of interpretation, with his own
commentary serving as a corrective to Porphyry’s. There he draws the
reader ‘upward’ to the truest and most genuine interpretation—as he
himself puts it, ‘leading the simples to the Ten, as did Archytas, with
whom Plato was in agreement’ (a b ±ºA N Œ Æ ı ªÆª ƒ
—ıŁÆªæ ØØ ‰ æåÆ, fiz ŒÆd —º ø ıªªª , 13,22).
Although Iamblichus’ approach appears to be a dramatic change of
style and emphasis in the treatment of the Categories, it would also go

1
On the theory that I sketched in Griffin 2009: chs. 1, 7.
2
As I also argued in Griffin 2009: ch. 1.
208 Conclusions
on to contribute to the later Neoplatonic respect for the ‘usefulness’ of
the Organon. And finally Simplicius, drawing Iamblichus’ lofty and
noeric exegesis to a level ‘accessible to the multitude’ (3,6–7), came to
produce the exhaustive commentary that we now possess, in which
the stratigraphy above, or something like it, is still encoded.
Throughout this long exegetical history, the Categories maintained
the ‘pole position’ that Andronicus granted it at the outset of the
Aristotelian reading list, functioning as a beginner’s guide to the
æºÅł Ø and ØÆØ of the ten genera of predication that we all
share. Without the interest that Andronicus’ and Boethus’ interpret-
ation started, we may never have come to regard the Categories as
holding a crucial role in understanding Aristotle’s thought and meta-
physics. Thus Andronicus, regardless of the influence of his editorial
work, has been responsible for a major factor in the development of
subsequent Aristotelianism.
Appendices
 APPENDIX 1

Persons and Sources

I. Main Persons Treated

Andronicus of Rhodes (fl. after 43 bce?). Ch. 3. A first-century bce Peripat-
etic, said by some later sources to have been scholarch in Athens, who
is widely credited with the publication of Aristotle that reintroduced
his esoteric works (the school treatises which we now possess, in contrast
to the popular dialogues that Aristotle also published but are now lost) to the
contemporary public. Among his works was a seminal catalogue of Aristotle,
and a paraphrasis of the Categories. Several unorthodox views of Andronicus
about the text have come down to us: for example, his replacement of the
categories ‘When’ and ‘Where’ with those of ‘Time’ and ‘Place’.
Eudorus of Alexandria (fl. c.50 bce?). An ‘Academic’ (on the use of the
term here, see Bonazzi 2013) and Neo-Pythagorean who discussed the
Categories in the first century bce, perhaps prior to or contemporary with
Andronicus’ paraphrasis. He may have introduced the criticism of the
Categories as an inadequate account of being, and specifically ‘intelligible’
being. At the same time, Eudorus may have found in the Categories a
congenial analogy for certain aspects of (Neo-Pythagorean) metaphysics.
The emphasis of his fragments suggests that Eudorus read the Categories
primarily as a work with something to say about the structure of being.
Ariston (fl. c.50 bce?). A rival of Eudorus in Alexandria, if he is to be
identified Ariston of Alexandria, who joined the Peripatos alongside Cratip-
pus of Pergamon. He shared many of Andronicus’ views, for example on the
definition of the Relative.
‘Archytas’1 (fl. c.50 bce?). The name given to the author of certain
influential Neo-Pythagorean pseudepigrapha, to signify their authorship by
the fourth-century Pythagorean and Academic Archytas of Tarentum; the
‘Pseudo-Archytan’ corpus is strongly influenced by Eudorus. At least three of
‘Archytas’’ texts dealt with the subject matter of the Categories, and the
Neoplatonic philosopher Iamblichus (see below) held that his works were
written prior to the Categories, and could profitably be employed in order to
obtain a higher understanding of the subject matter of the Categories.

1
See the introduction to T. A. Szlezák (1972) and M. Frede in Der Neue Pauly, as
well as Dörrie, Platonica minora, 300. I have also found the synopsis in Dillon (1977/
1996) and Bonazzi (2007) useful, with other literature given in chs. 3 and 3.1.
212 Main Persons Treated
Philo of Alexandria2 (20 bce to 50 ce). The great Jewish philosopher,
exegete, and polymath is mentioned briefly in ch. 3 and elsewhere, particu-
larly as a follower of Eudorus in some areas of his treatment of the Categories.
He is not separately treated as a major subject of this study, although there is
much of interest for future research in Philo’s use of the ten categories
(although not necessarily of the Categories) throughout his corpus.
Lucius (fl. later first century bce?). Apparently a Platonizing critic of the
Categories, perhaps datable before or contemporary with Boethus of Sidon;
Lucius’ fragments survive only from Simplicius On the Categories. He seems
to have broadly followed Eudorus, especially in his criticism of the Categories
as an inadequate treatment of intelligible being, and set the stage for the
discussion in Plotinus, Enn. 6.1. However, as I argue in ch. 4, there is very
little evidence for an historical figure Lucius, and the name may serve to
represent an entire Platonic–Pythagorean ‘school’ to which Boethus replied
under this name.
Boethus of Sidon (fl. first centuries bce–ce). An especially diligent and
influential Peripatetic reader of the Categories and (according to late sources)
disciple of Andronicus, who defended the text against the arguments of
Lucius, among others. Boethus sought to rebut the allegation that Aristotle
had failed to provide an adequate account of intelligible being, and is the first
scholar on record to ask instead how the Categories account of ousia can be
reconciled with that of the Metaphysics. He also reasserted the canonical
views of Aristotle in certain respects against the less orthodox positions of
Andronicus, for example, reinstating ‘When’ and ‘Where’ in the roster of ten
categories. Boethus’ formulation that the Categories concerns ‘simple verbal
expressions significant of beings, insofar as they are significant’ was later
adopted by Herminus—perhaps, as I suggest, in response to Athenodorus
and Cornutus—and became foundational for Neoplatonic and later seman-
tics. In general, I think that the texts I have attributed to Boethus unambigu-
ously belong to the Peripatetic Sidonian and not to the Stoic of that name.3
Athenodorus [of Tarsus?] (fl. c.50 bce or earlier?). A Stoic philosopher,
perhaps to be identified with the mentor of Octavian, who criticized the
Categories as dealing inadequately with the parts of speech.
L. Annaeus Cornutus (fl. c.60 ce). Cornutus is chiefly known to the
historical tradition as the tutor of the silver-age poet Persius and as an
associate or instructor of the emperor Nero. Among other subjects, he

2
On Philo of Alexandria, see Runia in Der Neue Pauly [I 12], with extensive
bibliography. On his use of Peripatetic material, see Gottschalk, 1141 and 1145–6.
3
In my original approach to Boethus (Griffin 2009: ch. 5), I exclude 10 of the 11
fragments in SVF under the name of Boethus Stoicus, which clearly stand apart from
the fragments collected above as representative of Boethus Peripateticus. An ambigu-
ous passage is Boethus Stoicus fr. 11 = Simplic. in De An. 247,24.
 Persons and Sources 213
wrote extensively on grammar and Stoic allegory (the latter work is extant),
and he composed a work of commentary that challenged both Athenodorus
and Aristotle on the Categories. Cornutus may, as I suggest, have defended
the Categories against the attacks of Athenodorus by pointing out that it
should be viewed as a contribution to the second ‘part’ of Stoic dialectic,
rather than the first (linguistic) part; this defence was then taken up by
Herminus and combined with Boethus’ formulation of the skopos.
Plutarch of Chaeronea (46 to c.122 ce). The prodigious writer Plutarch is
a helpful philosophical source for Eudorus of Alexandria (in Proc. An.,
discussed by Dillon 1977: 226 and by Karamanolis 2006: 124-5) and as an
historical source for Andronicus of Rhodes. A remarkable passage in his
Generation of the Soul in the Timaeus, which refers to Eudorus frequently,
suggests that Plato anticipated the full doctrine of Categories in the Timaeus,
and in fact has a more complete and accurate picture of it than does Aristotle;
this is also the view taken by Eudorus and (implicitly) ‘Archytas’, but it
survives explicitly in Plutarch.
Nicostratus (early second century ce). Like Lucius, whose views he
endorses, the fragments of Nicostratus survive solely in Simplicius On the
Categories. Praechter identifies him persuasively with the Nicostratus hon-
oured in an inscription at Delphi (Sylloge II3 Nr. 868). He shares Lucius’
concern with the adequacy of the Categories as an account of intelligible
being, and it is likely that Plotinus found these difficulties expressed in
Nicostratus, directly or indirectly. But Nicostratus also demonstrates (either
through his own interests or through the accidents of preservation) a stron-
ger interest in the rhetorical coherence and internal consistency of the
Categories, and appears to be equally concerned with these problems as he
is with matters of doctrine.
Achaïcus and Sotion (later first to early second century ce?): see Moraux
1984: 211. Not dealt with in a separate chapter. Scattered comments on the
Categories are ascribed to each, although we do not know the form that their
treatments took. According to Simplicius, both A. and S. held that we should
only speak of ‘relatives’ in the plural, never in the singular (in Cat. 159,25);
notably, they criticize ‘the ancient commentators’, namely ‘Boethus, Ariston,
Andronicus, Eudorus, and Athenodorus’, for failing to make the distinction.
Later, Achaïchus also appears to respond to Andronicus and Boethus regard-
ing the definition of the Relative (203,1). Other comments of Achaïchus and
Sotion are noted by Simplicius. Their treatment of earlier sources might
suggest that A. and S. played some role in conveying the views of the ‘ancient’
commentators; see also Porphyry in Cat. 111,22 and Ammonius in Cat.
66,14. Both A. and S. are only briefly treated here, and are not considered
in our discussion of the skopos of the text.
214 Main Persons Treated
Alcinous4 (second century ce). The author of the Didaskalikos follows the
earlier Platonizing tradition of claiming the categories for Plato—that is,
Plato anticipated Aristotle in his account of the ten categories. This syncre-
tizing and Platonizing attitude is common to this text as well as Eudorus,
‘Archytas’, and Plutarch.
Atticus (later second century ce). By contrast, the Platonist Atticus wrote
a polemical tract ‘Against those who interpret Plato’s philosophy by means of
Aristotle’, some of which is preserved by Eusebius (Praep. Ev. bks 11 and 15).
In taking this line, Atticus appears to follow L. Calvinius Taurus (second
quarter of the second century), who wrote ‘on the doctrinal differences’ of
Plato and Aristotle (Suda T 166, v. iv p. 509.12 Adler).5 Both mark a more
polemical stance toward Aristotle.
Aspasius (early second century ce). Aspasius—who taught Herminus, who
in turn lectured to Alexander of Aphrodisias—authored a series of commen-
taries, on the Categories, De Int., Physics, De Caelo, Metaphysics, and Nico-
machean Ethics. Of these, just the last is (in significant portions) extant today.
For our purposes, he is a key source for the ‘pathology’ of Andronicus and
Boethus, for which see the discussion in their chapters above.
Adrastus of Aphrodisias (early second century ce). We know from Galen
(De libr. propr. 42,20–43,1) that Adrastus wrote a commentary on the
Categories, and from Simplicius (in Phys. 122,33) that he wrote a commen-
tary on the Physics. An important position associated with Adrastus is the
decision to place the Topics immediately after the Categories at the head of
the Aristotelian corpus; he appears to have defined their relationship in a
work entitled ‘Concerning the Order of Aristotle’s Treatises’ (Simplic. in Cat.
16,1–4). He also revived the title ‘Before the Topics’ for the Categories—this
may suggest that he was responding to Andronicus, who explicitly rejected
the placement of the Categories (with the post-praedicamenta) before the
Topics, and with it, the title Ta pro tôn topôn. Adrastus notably discussed a
second version of the Categories, which the later commentators, following
Porphyry, dismiss.6 Aside from his work on the Organon, Adrastus also did
work in astronomy (cf. Sorabji in Sorabji and Sharples 2007: 579–80).
Herminus (later second century ce). Herminus is perhaps best known
today as a lecturer to Alexander of Aphrodisias. He has not attracted the
most charitable reviews from modern scholars; already Prantl (1855) judged
him an ‘offenbar höchst bornierter Mensch’, and Moraux (1983: 363–4) finds

4
See Dörrie and Baltes (III, 2002), 341; Dillon (1993); Whittaker in ANRW II 36 1;
and bibliography in Der Neue Pauly for Albinus and Alcinous.
5
See Dillon (1977: 237 ff.), Dorrie and Baltes, 310–23, and Praechter 101 ff.
6
On A. cf. Simplic. in Phys. 4.11, 6.4 ff; in Cat. 15.36, 18.16 ff.; Anon Prol in Cat.
32b36 Brandis; cf. Zeller III 1, 809 n. 3; Moraux Listes 58 ff. See recently Sorabji
(2007b).
 Persons and Sources 215
him almost wholly derivative of Boethus. I tried (2009) to suggest that
Herminus’ importance for the tradition on the Categories has been under-
estimated. He is the first on record to claim that the Categories should come
first in the Aristotelian curriculum for pedagogical reasons. He is, as I suggest,
the forefather of Porphyry’s influential theory of ‘double imposition’; he
moulds the ‘narrative’ of the historical dialogue regarding the Categories
that has come down to us, pitting Boethus of Sidon against the partisans of
the Categories as concerning ‘being’, and the partisans of the Categories as
concerning ‘language’. In so doing, he succeeds in revitalizing Boethus’
formulation of the subject matter of the work, and offers a theory of
signification (although we can only reconstruct bits and pieces of this)
which deploys a Stoic or Epicurean concept of ‘natural’ significance to
displace the mediating role of concepts (noêmata) found in the opening
lines of the De Int., which he broadly rejects. I also suggested that Herminus
left (so far as we can tell) no written works to posterity, and that Porphyry
was reliant on Alexander’s lost in Cat. for Herminus’ views. In connection
with this, I proposed (Griffin 2009: ch. 6) that there is some cause to
reconsider the identification of the Peripatetic Herminus with a Stoic Her-
minus mentioned by Longinus (preserved by Porphyry, Vita Plotini).
Sosigenes7 (later second century ce). Sosigenes, who also lectured to
Alexander, raised an important question for the Categories, inquiring what
is meant by ‘what is said’ (to legomenon): ‘word’ (phônê), ‘thought’ (noêma),
or ‘object’ (pragma) (cf. Simplic. in Cat. 7,4–9,24; Dexippus 7,4). His ‘tripar-
tite’ approach contrasts with the ‘bipartite’ approach of Herminus (who
considers only ‘words’ and ‘objects’). As we shall find, in the discussion of
ch. 1 and broadly in ch. 7, this inquiry would have a far-reaching influence.
Sosigenes himself did not choose between the three arguments (Dexipp. 7,4;
cf. Moraux, Hermes 95 (1967), 169 on Alexander).
Aristoteles (later second century ce). Perhaps teacher of Alexander of
Aphrodisias, as argued for by Moraux8 (but see now Opsomer and Sharples,
2000); he is mentioned by Galen as ‘a man preeminent in Peripatetic
philosophy’; Syrianus also mentions him as ‘the younger Aristotle, inter-
preter of Aristotle the philosopher’ (Galen Peri eth. 11,4 Müller; Syrianus in
Met. 100,6).
Alexander of Aigai. On one occasion Simplicius ascribes to this Alexander
the same view as Alexander of Aphrodisias about the skopos of the Categories
(in Cat. 10,19–20), and at in Cat. 13,11–18 again refers to the view of
‘the Alexanders’ (cf. Moraux II, 222). Simplicius at in de Cael. 430,32 states

7
On him see Zeller III 1, 813; Rehm RE III A 1, and Moraux.
8
He is apparently mentioned in Simplicius in de cael. 153,16; Alex. De an. II
(Mantissa) 110,4; Moraux discusses him in ‘Aristoteles, der Lehrer des Al. v. Aphr.’,
Arch. Gesch. Philos. 49 (1967), 169–82. See also P. Accattino, ‘Alessandro di Afrodisia
e Aristotele di Mitelene’, Elenchos 6 (1985), 67–74, who suggests (73) that Alex in Met.
166,19–67,1 follows one of Aristoteles’ lost discussions.
216 Main Sources
that the Aphrodisian Alexander cited Alexander of Aigai on Aristotle’s
argument at de Cael. 2.6, 288b22; see Moraux II, 223–5.

II. Main Sources

A much more comprehensive bibliography on the later ancient commenta-
tors may now be found in the Sourcebook produced by Richard Sorabji
(2005a–c, three volumes), and in Sellars (2004); likewise, a more compre-
hensive discussion of philosophers active during the first centuries bce may
be found in Sorabji and Sharples (2007). Most of the figures mentioned here
are also discussed in detail in Sorabji (ed.), Aristotle Transformed (1990).
Alexander of Aphrodisias (late second to third centuries ce). While
Alexander himself, the exemplary ‘commentator’ on Aristotle, is not system-
atically treated in the course of this study (with some exceptions, e.g. ch. 1
and 5), a number of his commentaries serve as sources. Unfortunately,
Alexander’s own commentary On the Categories is lost to us; but I suggest
in ch. 1 that some aspects of his discussion of the skopos, including his
criticism and development of his teachers Herminus and Sosigenes, may be
identified in the later sources. Several other works of Alexander are used in
the chapter on Boethus, in discussing the status of universals.
Galen (late second to third centuries ce). Like Alexander, the work of
Galen himself falls outside the scope of the current study. However, his texts
are cited in several places in this treatment.
Plotinus (c.204-70 ce), especially Enneads 6.1–3. The founder of the
philosophical tradition that we regard as Neoplatonism, Plotinus is here
chiefly referenced in the discussion of Lucius and Nicostratus, although he
does not cite them directly, and their fragments are drawn exclusively from
Simplicius.
Porphyry of Tyre (c.233–309 ce). Later sources rely mainly on Porphyry’s
lost commentary Ad Gedalium for their account of the Categories and its
earlier critics and defenders (cf. Smith 1993, Hadot 1990b, Chase 2003),
although Iamblichus also introduced other sources not used by Porphyry,
such as ‘Archytas’. A portion of Porphyry’s lost commentary may now have
been recovered in the Archimedes Palimpsest: see Chiaradonna et al. 2013.
(For Porphyry’s alleged ‘debate’ with Plotinus regarding the status of the
Categories, see my chapter on Lucius and Nicostratus, with recent bibliog-
raphy.) Porphyry’s lesser ‘introductory’ commentary in Cat. is used through-
out this study (text in CAG; translation and commentary in Strange, 1992), as
is his Isagoge (see the excellent introduction and detailed commentary in
Barnes, 2003). With respect to Porphyry’s logical influence, I have made
heavy use of Lloyd (1990) and several treatments by Sten Ebbesen (see
bibliography).
 Persons and Sources 217
Iamblichus (c.242-325 ce) According to Simplicius in Cat., Iamblichus
took over Porphyry’s commentary Ad Gedalium and (i) added an exegetical
approach called noêra theôria, a ‘higher criticism’ which reinterpreted the
Categories in the mould of Neoplatonic metaphysics (see my ch. 1 and Dillon
1997); (ii) adduced ‘Archytas’ for Neo-Pythagorean metaphysics, supporting
(i). The interpretation of the Categories as a text with something to say about
Platonic or Pythagorean metaphysics may be traced all the way back to
Eudorus (ch. 3.1), and Iamblichus evidently followed ‘Archytas’ in adopting
a similar view of the work. However, the earlier Neo-Pythagoreans such as
Eudorus and ‘Archytas’ (cf. 22,31 and 31,5 Thesleff) held that the Categories
concerned the sensible world only; Iamblichus seeks to show, as in several
examples discussed above, that the text can also lead the reader ‘upward’ to
the noetic realm. Iamblichus’ own commentary on the Categories is now lost,
but was heavily used by Dexippus (as Simplicius informs us), by Simplicius
himself, and by Olympiodorus.
Dexippus (mid-fourth century ce). A pupil of Iamblichus, Dexippus’ brief
surviving commentary in Cat. (text: CAG; translation and commentary in
Dillon, 1990) combines some of the defences of the Categories found in either
or both of Porphyry and Iamblichus (as Simplicius informs us). Dexippus
does not name names of sources as consistently as Simplicius, but he can be a
very useful guide and occasionally provides a fuller discussion of the argu-
ments used to refute a criticism. A prime example is his fuller discussion of
the views of Athenodorus and Cornutus (see ch. 6); although he does not
name them explicitly, the answering passage of Simplicius gives us their
identity, and Dexippus provides more detail. Dexippus can also be useful
to us for identifying where Iamblichus has added something to Porphyry’s
basic account.
Ammonius (c.435/45–517/26 ce). A disciple of Proclus of Lycia, the
towering Athenian Platonist of late antiquity (412–85 ce), Ammonius taught
the last great generation of Neoplatonist commentators—including Simpli-
cius, Olympiodorus, and Philoponus, among others. He was an influential
holder of a chair in pagan philosophy at Alexandria at a time when such a
position was rather controversial (cf. Westerink 1990). One commentary in
De Int. is known to survive from his pen; a commentary in Cat. constitutes a
student’s notes, as does the commentary On Prior Analytics I. I have made
some use of all of these commentaries in the following chapters. See Hadot
(1991) on the Alexandrian introductions to Aristotle, and see Westerink
(1990) and Wildberg (1990) on Ammonius.
Boethius (c.480–524/5 ce). Boethius followed Marius Victorinus in trans-
lating Greek philosophers into Latin. I have made considerable use of his De
Divisione in discussion of Andronicus (see ch. 2, where I discuss the likeli-
hood that the proem and conclusion are ‘Andronican’), and I have also made
218 Main Sources
use of his commentaries Peri Hermeneias and in Cat. (and to a lesser extent
his commentary on the Isagoge), particularly in ch. 1.
Simplicius of Cilicia (c.490–c.560 ce). As his commentary in Phys. is
invaluable for the recovery of Presocratic fragments, the commentary in Cat.
by the Neoplatonist Simplicius is the single most valuable source for the
earlier tradition on the Categories that reaches him from Porphyry and
Iamblichus. As we found above, he even informs us of the process that he
uses in comparing and reporting earlier commentaries now lost to us, which
is a great aid in evaluating his sources. And unlike many of the other later
commentators, such as Ammonius, Simplicius explores and reports the
views of earlier figures in detail, often with their names attached. We are
entirely reliant on Simplicius for the supposed existence, let alone the
positions, of some of the commentators who appear in his pages. The CAG
text of Simplicius in Cat. has been translated into English and annotated in
four volumes in the Ancient Commentators on Aristotle series (1–4 in Chase
2003; 5–6 in Fleet 2001; 7–8 in Fleet 2002; 9–15 in Gaskin 2000). I have
found the most recent introductory material and notes in Chase (2003)
especially exemplary and useful. The Categories commentary has also been
closely studied and translated into French by Hoffman and Hadot (1990),
with valuable comments on the problem of the skopos. The study given in the
next chapter surveys an example of the ‘stratigraphy’ in Simplicius’ text.
Olympiodorus of Alexandria (c.495–570 ce). A student of Ammonius,
Olympiodorus the Younger was among the last pagan philosophers to teach
at the Alexandrian school in the sixth century ce; he may, in fact, have been
the last pagan philosopher to hold the chair. Wildberg (2008b) and
Westerink (1990) are both useful introductions to Olympiodorus’ life and
times, with bibliography. His Prolegomena to Aristotle’s Logic and Commen-
tary on Aristotle’s Categories (Busse’s text in CAG) are used extensively in my
discussion of the skopos of the treatise. Again, see Hadot (1991) on the
Alexandrian introductions to Aristotle in general. I have translated some
excerpts of Olympiodorus in Cat. in ch. 1.
John Philoponus (c.490–570 ce). Philoponus was also a pupil of
Ammonius—although, unlike Olympiodorus, he did not succeed Ammonius
in a chair. A Christian philosopher who brought innovation to the commen-
tary form elsewhere, his commentary in Cat. was relatively orthodox (cf.
Hadot 1991); I have made several references to it, but my case studies of the
Alexandrians following Ammonius have chiefly focused on Olympiodorus.
David (Elias) (later sixth century ce). The writings on the Categories
ascribed to ‘David’ mention Olympiodorus several times, and their author
may have been his student. They may have been written during Olympio-
dorus’ life, or later (cf. Westerink 1990; Wildberg 1990; Wildberg 2008b).
 APPENDIX 2

Andronicus’ Publication and Works

I. Publication and Catalogue of the Aristotelian Corpus

A broader survey of the general problems associated with Andronicus of
Rhodes’ alleged ‘Roman edition’ of Aristotle may provide a useful context for
the questions associated with the Categories. This appendix is divided into
two sections: the first will briefly survey the sources for Andronicus’ ‘publi-
cation’ and pinakes, with critical discussion, and the second will offer an
assessment of the evidence within the narrower confines of the reception of
the Categories.1 (Subsequently, I briefly review evidence for some of Andro-
nicus’ other important works.)

Sources and Discussion

Despite Barnes’ salutary reminder of the poverty of the direct evidence, the
‘legend’ of Andronicus’ formative collection, publication, and organization of
Aristotle and Theophrastus must have begun somewhere. Like so much of
our information about the first-century Peripatetics, this particular story may
have gained currency with Porphyry (Vita Plotini 24), who tells us that
‘Andronicus, the Peripatetic, divided the works of Aristotle and Theophrastus
into treatises (
æƪÆÆ), collecting related material (a NŒÆ
Ł Ø)
into the same [place]’, and that Porphyry himself followed suit in his edition of
Plotinus. Barnes, reasonably enough, keeps Porphyry waiting ‘in the wings’
while he tackles the earlier sources such as Strabo and Plutarch, but he also
introduces Vita Plotini 24 as ‘the text upon which Andronicus’ reputation as the
founder of modern Aristotelian scholarship ultimately rests’ (37).
It takes at least three separate but overlapping pieces of testimony to
establish the ‘orthodox’ narrative of Andronicus as the individual responsible
for the first reliable publication of Aristotle’s school works, and, by extension,
as the individual who deserves credit for the first-century revitalization of the
Peripatos. These are Porphyry’s evidence (VP 24.7), coupled with Strabo’s
report that the Hellenistic Peripatos lacked Aristotle’s school treatises until

1
Here I reiterate my conclusions in Griffin (2009). This ground has recently been
excellently covered by Hatzimichali (2013) and Chiaradonna (2013); I find myself in
broad agreement with them. Natali (2013: 101–4) offers a clear summary of the story;
see also Too 2010: 25–31.
220 Publication and Catalogue of the Aristotelian Corpus
the first century (13.1.54), when they were inundated by damaged and
inferior copies, and Plutarch’s addition (Sulla 26.1–2) that ‘the Rhodian
Andronicus, gaining ready access from Tyrannio to the copies, made
[them] public (N   ŁE ÆØ), and composed the catalogues that are
now in circulation (f F çæ ı
 ÆŒÆ).’ The later tradition took
the same line, presumably following Porphyry; but there is a reasonable
argument, as Tarán observes, for tracing the later Neoplatonic reports back
to the same brief textual notices now in our possession, namely Strabo,
Plutarch, and Porphyry, or something very much like them. These notices
might bear multiple interpretations.
There are several reasons not to dismiss Strabo’s tale of the Hellenistic
history of the ‘library’ itself as a fabrication. Some of its essentials are
independently ratified. Neleus did take possession of the books of Theo-
phrastus, according to the latter’s will (Diog. Laert. 5.53); and according to
Andronicus’ countryman Posidonius (ap. Athenaeum 5.214d, if the attribu-
tion is correct), Apellicon did purchase some ‘library of Aristotle’ several
centuries later. Moreover, the Arabic translation of Ptolemy’s catalogue of
Aristotle, which may be a kind of digest of Andronicus’ pinakes,2 refers to a
collection of books ‘found in the house of a man called Apellicon’, perhaps,
with Barnes 1997, as a section title.
Posidonius, as preserved by Athenaeus, may be the source of the entire
story reported by Strabo and Plutarch. On the other hand, Athenaus himself
claims elsewhere that Neleus sold his books to Ptolemy for the Library at
Alexandria (1.3a),3 which is difficult to reconcile with the ‘Posidonian’ tale.
Perhaps, as Gottschalk suggests (1987: 1085), Athenaeus tried to harmonize
Strabo’s story with a different tradition that recalled Ptolemy purchasing the
works of Aristotle and Theophrastus in Rhodes and Athens, in a way that
would maintain the pre-eminence of Alexandria itself. Alternatively, there
may have been two ‘libraries’ in play. Neleus, perhaps, sold Aristotle’s Ø ºÆ
to Ptolemy, but kept Aristotle’s own writings aside; or perhaps it was
Aristotle’s ‘exoterica’ that circulated, while Neleus managed to maintain a
portion of the collection, comprising the unpublished ‘esoteric’ texts that
allegedly vanished into the underground of Skepsis.4 (Admittedly this is not
directly supported by either source.)

2
The Arabic text itself suggests that the author composed the pinakes from
memory, as his copy of Andronicus was not to hand—from which it is reasonable,
though not necessary, to infer that the author’s catalogue is loosely based on Andro-
nicus’ own.
3
cf. H. B. Gottschalk, ‘Notes on the Wills of the Peripatetic Scholarchs’, Hermes
100 (1972), 335–42.
4
So Schweigäuser on Athenaeus, Stahr, Aristotelia II, 55 and following;
O. Regenbogen, Theophrastus v. Eresos, RE Suppl. 7 (1940), 1377 and following;
I. Düring, ‘Ariston or Hermippus? A Note on the Catalogue of Aristotle’s Writings,
Diog. L. v. 22’, Classica et Medievalia 17 (1956), 13.
 Andronicus’ Publication and Works 221
In either case, it may strain the bounds of credibility that all of the original
texts of Aristotle’s lectures were lost, as Strabo indicates, and that the
intellectual treasury of the Lyceum was borne away in a single copy. The
‘letter of Aristotle to Alexander’ preserved by Aulus Gellius (20.5; cf. Plezia
1961, 42 and 127), presumably circulated by Andronicus himself among the
‘letters of Aristotle’, suggests that someone thought the esoteric works were
circulating during Aristotle’s lifetime, and that Andronicus credited this
report with authenticity.5 Stahr’s and Zeller’s collections of the evidence for
Hellenistic reading of the Aristotelian works,6 coupled with Barnes’ careful
survey (1997), agree that several of the school treatises, including the books
now collected as the Analytics, were available to read in several locations
under the Hellenistic kingdoms, including Rhodes. Hermippus, working at
Alexandria, had enough material to compile a catalogue of Theophrastus’
oeuvre (see fr. 54 Wehrli). Diogenes Laertius also records the contents of a
Hellenistic catalogue of Aristotle including logic, rhetoric and ethics (5.22–7,
and 42–50). Notably that earlier catalogue demonstrates a different organ-
ization than our ‘standard’ edition.7
About what information, then, can we be confident? (1) It is reasonably
uncontroversial that a copy of the works of Aristotle and Theophrastus
passed to Neleus, as Strabo suggests, bolstered by the circumstantial support
of Theophrastus’ reported will; this should have happened before the 270s
bce. (2) It is also uncontroversial that a copy of the works of the school came
down to an Apellicon of Teos in Athens, sometime around 90 bce.
(3) Finally, it is reasonably uncontroversial that the Rhodian Andronicus
published a copy of the works of the school some decades later in the first
century bce, accompanied (presumably, though not necessarily, at the same
time) by a significant catalogue (
 ƌ) that had displaced its Hellenistic
predecessor or predecessors by the lifetime of Plutarch, and would win the
respect of Porphyry in the third century.
It is on the interstices between these three events that criticism centres.
There are no obvious reasons to doubt the reports themselves—the will of
Theophrastus bequeathing books to Neleus, the purchase of the books of
Neleus for Alexandria, the sale of an Aristotelian library to Apellicon in the
first century—but perhaps a fictional fabric of continuity has been woven to

5
cf. Gottschalk (1088) and M. Plezia, Aristotelis Epistularum fragmenta cum
Testamento (Warsaw, 1961), 42f., 127 ff. Of course, as Peter Adamson has kindly
pointed out to me, we can hardly rule out the possibility that Andronicus invented
some of the letters himself.
6
Stahr, Aristotelia II, 92 and following, and Zeller, II.2, 144 and following, with
references, collect the evidence for Hellenistic citations of the treatises; Gottschalk
(1085) points out that many of those held to have consulted the school treatises were
old enough to have done so at the Lyceum before their removal to Skepsis.
7
See P. Moraux, Les listes anciennes des ouvrages d’Aristote (Louvain, 1951), 313.
222 Publication and Catalogue of the Aristotelian Corpus
relate them, positing that all three report the history of a single collection of
books, the autograph of the first Peripatetic Masters. Such a fabric would, after
all, make the story far more interesting, and provide narrative continuity. On
the assumption that a single library was involved, some original source of
Strabo’s story may have asked: what happened during the long gap between
Neleus’ receipt of the books and their appearance in the hands of Apellicon of
Teos two centuries later? If one knew something of Neleus’ reported family
history, or was aware that the texts purchased by Apellicon were full of holes,
or had some polemical axe to grind against the Attalids, one could easily come
up with a clever story such as the romantic ‘tunnel’ of Strabo, where the scrolls
were hidden away to protect them from prying Pergamene eyes. On the other
hand, if one were keen to promote the scholarly prestige of Alexandria, as
Gottschalk (1987) suggests that Athenaeus was, one might well imply that it
was these very books, the autographs of Aristotle himself, that were purchased
by Ptolemy from Neleus and employed to produce the well-known Alexan-
drian catalogues of Aristotle and Theophrastus.
The catalogue of Ptolemy, which now survives only in the Arabic version
ascribed to ‘Ptolemy the unknown’, lends some report to the story. At no. 92,
AB p. 230 (see Düring 1957, incl. commentary p. 245), Andronicus himself—
again, on the assumption that the digest of Ptolemy is a decent representation
of Andronicus’ catalogue—appears to include a heading ‘Books found in the
library of a man called Apellicon’.8 If Strabo’s source constructed a romance in
which Apellicon’s mysterious scrolls represented, not just a few tattered pages
of dubious provenance, but the long-lost autograph of the Master, then
Andronicus’ possession of copies of these same texts would serve to explain
the superiority of his publication. In reality, one might suggest, Andronicus
was a capable Aristotelian scholar and philologian who collated many extant
copies from around the Mediterranean, including copies of works then
regarded as not particularly interesting, or mingled with spuria, and produced
a text which came to be accepted as superior in the fullness of the next century.
But this, it seems to me, would be to grant that it really was the quality of the
texts, particularly in the modern sense of an editio princeps, that got Andro-
nicus’ generation and its successsors excited about his publication; in reality,
what evidence we have suggests that the later tradition, including Porphyry
and the second-century Peripatetic Herminus, found Andronican readings to
be less authoritative than what they read as the paradosis (cf. Porphyry ap.
Simplicium, in Cat. 30,3; Simplicius in Phys. 440,13).

8
On Düring’s view, this embraces certain items (93–9), which are not particularly
interesting and not philosophical; the heading may refer to books found only in
Apellicon’s library.
 Andronicus’ Publication and Works 223
In fact, the entire story of Sulla’s capture of the texts from Athens could be
a fabrication. Our fabricator, perhaps, knew that Andronicus collaborated
with the grammarian Tyrannio in Rome before producing his edition. How
did they get hold of the ‘library of Aristotle’? Apellicon was in Athens; the
great general Sulla sacked Athens shortly after he procured the collection,
and brought spoils back to Rome; any reader of Cicero knows about Sulla’s
library; Sulla is the obvious candidate to have taken the books to Rome, and
his library the ideal place for them to have languished for several decades
longer. If this particular part of the story was mere romantic invention, it
would readily explain why Cicero never mentioned these seminal tomes
lurking in his friend’s library; perhaps they were never there at all. For that
matter, it would readily explain why Strabo never mentions Andronicus in
connection with his Skepsis story. Like the rest of the story, there are several
genuine events that have been linked together by bright imagination. Such a
line of criticism would father such inventions on Strabo’s and Plutarch’s
sources. In fact, Plutarch was accused of some such invention by Diels and
Susemihl, focusing on Andronicus’ work in Rome.9 Perhaps the best testi-
mony against such a sceptical line is the mention of Apellicon in the
catalogue of Ptolemy the Unknown, but this too is susceptible to scepticism.
Was there anything original about Andronicus’ publication? Other testi-
mony seems to imply that the esoterica were available in some form before
Andronicus—for example, Philodemus provides fairly good reasons to
believe that Epicurus read some Analytics of Aristotle,10 and Andronicus
himself preserves the so-called ‘letter of Alexander’ claiming that Aristotle
had published the Metaphysics in some form. If Andronicus held the letter to
be authentic, presumably he himself could not have claimed originality for
his publication of the Metaphysics. Barnes (1997: 50–63) demonstrates the
difficulty of relying on the structure of various texts to show the importance
of the Andronican publication; there is no evidence, Barnes argues, that
Andronicus played a major role in the formation of the Physics (cf. Simplic.
924,18, 1036,18, on the seventh book); it appears that a Rhetoric like ours was
available before Andronicus, and Cicero writes of Antonius reading 
Ø in
the nineties bc (de Orat. 2.36.152); Cicero himself clearly knew the Topics
(Topica 1.1) in some form, possibly directly.

Assessment of Andronicus’ Role

Interest in Andronicus’ originality as an editor (in the sense of producing
something like an editio princeps by collating manuscripts) may be a modern

9
H. Diels (1879) and F. Susemihl, Die Lebenszeit des Andronikos von Rhodos,
Neue Jahrb. F.cl. Philologie 151 (1895), 226 ff.; cf. Moraux I 48 f.
10
Philodemus in P. Herc. 1005, fr. 111.
224 Publication and Catalogue of the Aristotelian Corpus
invention,11 not unreasonably drawn from Strabo and Plutarch. The ancient
tradition demonstrates no special interest in Andronicus’ activity as a textual
critic. (On the contrary, Porphyry and Simplicius tend to brush off ‘Andro-
nican’ readings that are contrary to their paradosis.)12 An example of modern
embellishment might perhaps be spotted in the Loeb translation of one key
passage in Plutarch: ‘The greater part of the collection passed through the hands
of Tyrannio the grammarian, and Andronicus the Rhodian, having through his
means the command of numerous copies (P
æ Æ Æ H I تæçø ), made
the treatises public.’ But the Greek P
æ Æ Æ H I تæçø does not need
to say anything about the numerousness of copies available, which, in the
English translation, might lead us to infer that the availability of multiple
manuscripts was somehow relevant to the quality of the publication. Rather,
a I ªæÆçÆ are presumably exactly those copies made from Tyrannio’s
single set of (tattered) originals, which were just mentioned; and P
æ Æ Æ
could refer merely to the fact that Andronicus possessed ready and convenient
access to those copies (unlike earlier Peripatetics), not that he had a great
quantity of copies to compare. Barnes’ careful critique of the view that Andro-
nicus produced an authoritative text, then, certainly removes a modern layer of
interpretation, by showing that Andronicus’ enterprise did not necessarily
presage that of Immanuel Bekker. For Plutarch in Sulla 26, it is the information
that Andronicus made the texts widely available (N   ŁE ÆØ) and that his
catalogue became current (f F çæ ı
 ÆŒÆ) that deserves atten-
tion, not the readings in the text and how he arrived at them.
Porphyry’s testimony in the Life of Plotinus (24.7) provides more detailed
and direct information about Andronicus’ process. Thus from Porphyry’s
famous observation that Andronicus a æØ  ºı ŒÆd ¨çæ ı N

æƪÆÆ ØEº a NŒÆ
Ł Ø N ÆPe ı ƪƪ , and that
Porphyry himself (at least partly) followed Andronicus’ model in his own
publication of Plotinus’ Enneads, it has sometimes been inferred that
Andronicus was responsible for ‘pulling together’ the books of, say, the
Metaphysics or the Topics, or the Physics into their current collected form
as free-standing works, perhaps adding cross references and bridge passages
for coherence. But Barnes has effectively undermined that hypothesis as well,13

11
Hatzimichali offers a very useful discussion of the notions of editorial activity
that could be applied in antiquity (2013: 1, 19–23).
12
For example, Porphyry seems to have brushed off the authority of Andronicus’
text in a critical passage of the Categories, preferring the authority of the second-
century Peripatetic Herminus ‘and just about everyone else’ (in Cat. 30,3). Simplicius
in Phys. 440,13 notes an Andronican reading (if that is what the comment amounts
to) as a curiosity, upon which Andronicus built an innovative but unusual interpret-
ation of one lemma.
13
Barnes also suggests that Porphyry’s own method was not to ‘invent’ treatises as
Andronicus is supposed to have ‘invented’ the Metaphysics by
Ł Ø N ÆPe
 Andronicus’ Publication and Works 225
which similarly seems to be a relatively modern layer on the tale. It is certainly
not self-evident that this is what Porphyry means. What Porphyry himself did
in the Enneads was not, in any case, to invent treatises by pulling together
previously unrelated material and adding bridge passages or cross references,
but simply (as he himself claims) to pull self-contained treatises into the same
collection, wherever he believed they contained similar subject matter, and
within each collection to organize the treatises in such a way that easier or
‘lighter’ material came first (Œ fiÅ b K Ø a NŒEÆ ç æø ı çæÅ Æ
f ŒÆd Ø
æÅ E KºÆçæ æØ
æ ºÆ Ø , 24.14–16).
This final point, I think, deserves somewhat greater emphasis: giving Ø

æÅ E KºÆçæ æØ
æ ºÆ Ø , the starting position to the lighter
material, is also the reasoning that Porphyry endorses in explaining the
location of the Categories at the outset of the Aristotelian curriculum (in
Cat. 59,21–2), and the argument that Porphyry provides for the introductory
nature of his own Isagoge (1,8–9) as a precursor to the study of categories (N
c H
Ææa æØ  ºØ ŒÆŪæØH ØÆ ŒÆºÆ , 1,2–3). As Porphyry
says that he followed this principle on Andronicus’ authority, and Porphyry
himself applies it to the Categories, we might speculate that Andronicus took
a similar line in the organization of the Aristotelian corpus, and that his
argument for the location of the Categories in the reading of Aristotle
provided impetus to this tradition. Indeed, we are told elsewhere that
Andronicus argued for the introductory role of logic (Philop. in Cat.
5,15–20), although we do not know exactly how he argued.
For our purposes, then, we can leave aside the attractive touches of
Andronicus the textual critic and the compositor of the Metaphysics, and
focus on his status as a publisher and a cataloguer, but perhaps most
importantly an organizer, of the corpus. The primary texts do seem to permit
some inferences about the influence of an Andronican ‘publication’, but that
influence appears to have had (relatively) little to do with the readings that
Andronicus chose, or with the internal structuring of pragmateiai made up
of multiple books, like the Topics and Metaphysics. Rather, what seems to
have been most influential is the order in which Andronicus presented the
works of Aristotle, and the arguments that Andronicus made in favour of
that order.

ı ƪƪ . (Of course, in a different sense that answers to the other half of Andro-
nicus’ alleged activity, N
æƪÆÆ ØEº, one might argue that Porphyry did
invent treatises.) Thus, for example, Porphyry seems to have split the longer treatises
of Plotinus such as On the Soul, and indeed 6.1–3 On the Genera of Being, into
separate books—perhaps in order to arrive at the number of fifty-four individual
works. Perhaps we should understand that Andronicus found pragmateiai with
divisions that struck him as nonsensical, and organized them to have more rational
partitions.
226 Publication and Catalogue of the Aristotelian Corpus
Andronicus certainly published a catalogue (
 ƌ) of the works of
Aristotle and Theophrastus. We can, I think, take a confident line not merely
about the existence of this catalogue, but about its material influence. It was
‘now current’ on the cusp of the second century (Plutarch, Sulla 26.1–2), and
this is just the kind of point on which a polymathic, well-travelled scholar like
Plutarch should be able to advise us. More than a century later, Andronicus’
stylistic example was followed by Porphyry—and presented, as the VP passage
implies, in order to underwrite and justify Porphyry’s own approach to the
Enneads. The passage of Ptolemy the Unknown,14 however else we may put it
to use, shows that a catalogue carrying Andronicus’ name possessed ‘canon-
ical’ status in Ptolemy’s own lifetime (whenever that might have been), and
that it was not like the Hellenistic catalogue preserved by Diogenes Laertius
(5.22–7)—which, notably, omits our Categories (under that title) altogether.15
More specifically, we can infer from later reports that Andronicus’ cata-
logue opened with the Organon, and that here, or elsewhere, Andronicus
provided some justifications for the position of ‘logic’ at the propylaea of
Aristotelian studies (discussed in ch. 2); this organization of the corpus was
also to prove influential. Indeed, it was one area where the later tradition,
including Porphyry and the Neoplatonists, favoured Andronicus’ view over
that of his ‘pupil’ Boethus (cf. Philop. in Cat. 5,15–20), who recommended
commencing the study of Aristotle with physics. As we have at least some
evidence that Andronicus felt the need to argue for the usefulness and import-
ance of logic in his effort to build an Aristotelian ‘system’ based on the corpus,
we may infer that this was not taken for granted, and that Andronicus played
something of an original role in asserting (or reasserting) the importance of
logic in Peripateticism, and of Peripatetic logic in general. This would be
supported by the evidence that Andronicus paid special attention to the
Categories, if this is not merely an artefact of the later sources.16

14
See the edition of the catalogue of Ptolemy by Hein (1985); I am indebted to
Stephen Menn for pointing me to this work. The name ‘Ptolemy the Unknown’ or ‘the
Stranger’ likely derives from the misunderstanding of —ºÆE   as
—ºÆE ˛ ; see Rashed 2005: ccvii, Hatzimichali 2013: 19 n. 57. The catalogue
had at least five books, if not more (see item 97 in Düring 1957: 230); Littig 1894:
18–25 offers a reconstruction, and Plezia 1946: 16–35 studies it in detail, as does
Moraux 1973: 58–94. For the importance of Diogenes’ catalogue, and the fact that it is
more than a mere ‘library list’, see Hatzimichali 2013: 23–4.
15
See Moraux 1951: 131, Düring 1957: 69, and Gottschalk 1987: 1103.
16
We are aware that Andronicus provided a ‘paraphrase’ of the Categories (so
Porphyry via Simplicius: in Cat. 26,17, 30,3), and we are not directly aware of any
other such ‘paraphrases’; this may merely be an accident of preservation, but in other
cases where Andronican ‘readings’ are cited, such as the Physics (cf. Simplic. in Phys.
440,13) it appears that Andronicus has offered a reading of the Aristotelian text rather
than an interpretive paraphrase. This too might suggest some special concern with
logic.
 Andronicus’ Publication and Works 227
It is sometimes pointed out—against the enduring influence of the ‘An-
dronican’ ordering of the corpus—that the De Interpretatione appears in the
later catalogues and discussions, but Andronicus athetized it (presumably, he
marked it in his catalogue as not written by Aristotle, with an explanatory
note; cf. Alexander in An. Pr. I 160,28–161,1; Ammonius in Int. 5,28, 7,13;
Philoponus in Cat. 27,18–26 and 45,8–12).17 Yet such an argument appears
to assume that Andronicus’ athetization entailed the deletion of the work
from his catalogue, a kind of damnatio memoriae. It seems much more likely
that the De Int. was accompanied by a marginal siglum in the Andronican
catalogue pointing the reader to discussion of Andronicus’ view that some-
one other than Aristotle had written the work (and in this discussion,
presumably, Andronicus explained how the author of the De Int. cross-
referenced a De Anima that did not appear to be the De Anima he published
under the name of Aristotle). We know that these pinakes of Andronicus
included titles, capita, and comments on authenticity; it is presumably here
that Andronicus questioned the De Interpretatione. Later scholars could
perfectly well continue to use Andronicus’ catalogue whilst ignoring (or
rebutting) his associated comments against its Aristotelian authorship.
Moreover, although Andronicus suspected that the De Interpretatione was
by some other philosopher, there is no evidence that he thought there was
something wrong with it, or held its contents to be somehow philosophically
suspect. He simply doubted that Aristotle wrote it. But Andronicus did not
think that Aristotle had all the answers when it came to Aristotelian logic.
We have multiple examples of Andronicus’ efforts to improve and innovate
on Aristotle’s ideas, especially in the Categories.18 And we have some evi-
dence that Andronicus sought to distinguish himself from the ‘earlier’
Peripatos in the field of logic;19 moreover, he held strong views about the

17
Cf. Moraux 1973: 117–19.
18
For example, he famously replaced Aristotle’s categories of Where and When by
Place and Time (Simplic. in Cat. 134,5, 342,23, 357,28). He added a fifth kind of
quality to Aristotle’s own list (263,19–22). He seems to have suggested—if this is not a
textual criticism but a philosophical point—that Aristotle could have expressed
himself more clearly at Cat. 6a37, as his extant definition of the relative is circular
(202,5). All of these innovations sparked defences of Aristotle by Andronicus’ suc-
cessors, especially Boethus. For Andronicus’ attitude to Aristotle more generally, see
also Moraux 1973: 99.
19
In the opening to Boethius’ De Div., Andronicus is cited as remarking on the
value and utilitas of logic, and (see discussion in Magee) he may be seen as the source
of criticism of ‘earlier Peripatetics’ for failing to differentiate accidents, on the one
hand, from genera and species, and differentiae, on the other hand. Indeed, through-
out Boethius’ De Div. everything is divided into a per se and secundum accidens
dichotomy, a point that may be fruitfully compared to Andronicus’ view that the ten
categories can be reduced to something like substance and accident.
228 Publication and Catalogue of the Aristotelian Corpus
way in which the Organon should be organized in order to present a coherent
and systematic picture of Peripatetic logic. Indeed, setting aside the obser-
vation that Andronicus believed someone other than Aristotle had written
the De Interpretatione, we have every reason to suppose that Andronicus is
the source, or at least the proximate source, of the ordering Categories–De
Int.–Prior Analytics–Posterior Analytics–Topics that appears in the catalogue
of Ptolemy and is assumed by the Neoplatonists. Similarly, we may plausibly
assume that Andronicus endorsed or originated the associated pedagogical
view that the Organon systematically builds from terms to propositions to
syllogisms to demonstration.20 (For the influence of this organization, per-
haps stemming from Stoicism, see Bodéüs 2001: xiv–xvi.) Andronicus felt
sufficiently strongly about this organization to argue against those who
would place the Categories before the Topics, a view which, he felt, had
caused chapters 10–15 to be appended (wrongly) to the Categories in the
first place (see Simplic. in Cat. 379,9–11).
All in all, Andronicus’ most lasting influence appears to reside in his
organization of the corpus Aristotelicum—that is, the order in which he
placed the works, and the account that he gave of why they belong in that
order. Perhaps we can press this further to include the account that he may
have given of how they produce a coherent system in that order. His views on
how the texts should be read relative to each other appear to have been far
more influential than his views about specific readings in any given text, or
even the correct attribution and authenticity of the De Interpretatione or the
post-praedicamenta to the Categories. Due to the emphasis that Porphyry,
and so the successors of Porphyry, placed on the Categories itself, we have the
best evidence for Andronicus’ approach to the Cat. as opposed to, say, the
Topics. Certainly some of his contemporaries seem to have followed his lead
in certain respects, as did Ariston and perhaps ‘Archytas’ on the redefinition
of relation (Simplicius in Cat. 202,5, 203,4–6), and Eudorus on the replace-
ment of Where and When by Place and Time (206,10 and following).21 It
may be difficult to be sure who is influencing whom. Nonetheless, these cases
may reflect Andronicus’ own interest in the Cat., if indeed he carried the
Categories to the head of the curriculum in order to create a systematic
progression.

20
Thus Andronicus seems to have argued, against some unnamed predecessor, that
the Categories should not come immediately before the Topics (cf. Simplic. in Cat.
379,9–11), and for reasons we discussed in ch. 2, he seems to have placed the Categories
at the outset of the curriculum.
21
With Szlezák (1972), we find that ‘Archytas’ relies on several points that (to our
knowledge) originate with Andronicus. For instance, he and Ariston both follow
Andronicus’ apparent redefinition of Relation (Simplicius 202,2). In other respects,
however, ‘Archytas’ differs from Andronicus: he does not place the category of
Relation last, but rather fourth.
 Andronicus’ Publication and Works 229
II. Other Works of Interest
Andronicus’ Book ‘On Division’
Quam magnos studiosis afferat fructus scientia diuidendi quam-
que apud peripateticam disciplinam semper haec fuerit in hon-
ore notitia, docet et Andronici, diligentissimi senis de diuisione
liber editus; et hic idem a Plotino grauissimo philosopho com-
probatus et in libri Platonis, qui Sophistes inscribitur commen-
tariis a Porphyrio repetitus, et ab eodem per hanc introductionis
laudata in Categorias utilitas. Dicit enim necessarium fore gen-
eris, speciei, differentiae, proprii, accidentisque peritiam, cum
propter alia multa tum propter utilitatem quae est maxima
partiendi. (Boethius, De Divisione, Proem)
Boethius’ monograph On Division22 begins with a short notice about ‘An-
dronicus the Peripatetic’, an ‘ancient’ and exceedingly ‘careful’ philosopher
(senis . . . diligentissimus) who also published (editus) a book On Division
(4,3–11). Boethius takes Andronicus as his authority for the usefulness of the
science he will expound (quam magnos studiosis afferat fructus scientia
diuidendi, 4,3) and for its enduring value in the ‘Peripatetic’ tradition
(quamque apud peripateticam disciplinam semper haec fuerit in honore
notitia). According to Boethius, Andronicus’ book was praised by Plotinus
and adapted by Porphyry in his own commentary on the Sophist; indeed,
Porphyry’s essay is almost surely the source for Boethius’ knowledge of
Andronicus and for the main contents of the De Div. (see Magee 1998,
Plezia 1946, Moraux 1973, Littig 1890, cited above).
Moreover, Littig (1890: iii.10) had suggested that Andronicus may be the
original model for the entire work, a view that Moraux and Magee have
argued against convincingly. Nonetheless, the proem is strongly suggestive
that Andronicus provided a model of sorts for Porphyry’s essay on division
in the latter’s Sophist commentary, while Boethius in turn used Poprhyry’s
essay to construct his own. And while we cannot be at all sure which details
in Boethius are Andronican (or pre-Porphyrian), the basic expression of
value for diairesis at the outset of the work should be attributed to him.
Moreover, when Boethius suggests that ‘earlier’ Peripatetics had failed to
make a sufficiently clear distinction between the two major modes of division

22
On which see the introduction of Magee, especially ‘Boethius, Porphyry and
Andronicus’, and notes to the prologue and final lines of the work. It is also studied by
Plezia 10-15, 44–6 and Moraux i.120–32. Littig (ii.12–15), followed by Plezia, argues
that Boethius’ Div. was not only an homage to, but also closely modelled on, the work
of Andronicus. In fact, if there appears to be Stoicism at work in the De Div.,
Andronicus may have been influenced by Posidonius (Littig iii.10).
230 Other Works of Interest
under discussion, per se and secundum accidens.23 He credits the correct
recognition of that contrast to ‘later’ Peripatetics (48,26–50,5 Magee), and
Magee makes a strong case (ad loc.) that these ‘later’ Peripatetics include
Andronicus. This might suggest that the central concept of the De Div., that
is, the strong emphasis on this bipartition of per se and secundum accidens,
does go back to Andronicus.24 The terminology of ‘earlier’ and ‘later’ Peri-
patetics may have particular relevance to Andronicus’ own role (compare
Aspasius in EN. 44,20 and following, where ‘earlier’ and ‘later’ seems to mean
pre-Andronican and post-Andronican, and Aspasius explicitly makes An-
dronicus and Boethus ‘later’).
That evidence might be compared to another fragment studied above
(Simplicius, in Cat. 63,22–8), where we learn that Andronicus believed the
ten categories might be reduced to just two groups, namely ‘by itself ’ (kath’
hauto) and ‘relative’ (pros ti). In that passage, Andronicus is (puzzlingly)
named alongside Xenocrates, an Old Academic, while Simplicius compares a
similar bipartition into substance and accident, respectively analogous to
kath’ hauto and pros ti. We also encounter various examples of Andronicus
pointing to problems caused (perhaps) by an apparent failure to sufficiently
distinguish genera, species, and differentiae, on the one hand, from accidents,
on the other (e.g., Simplic. in Cat. 54,8–21; cf. Chiaradonna et al. 2013:
177–9). Combining these testimonies, we might infer that Andronicus
made some comment to the effect that all ten modes of predication need to
be clearly distinguished into the kath’ hauto and pros ti, lest one fall into the
errors of the ‘earlier’ Peripatetics.
Possibly an immediate source for this division in Andronicus is Aristotle
himself, in passages like Met. 5.6, 1015b16 and 5.7, 1017a7, where he appears
to distinguish predication kath’ hauto from predication kata sumbebekos,
and points to the example of the ‘musical’ as something that can take on
another property kath’ allo. Plato in the Sophist (255C12–D7) is the ultimate
source for the twofold division of beings along these lines:25

23
It is on this ground that the book De Div. seeks to render an account of the
science of ‘division’. In a core passage (6,17–8,2), division is itself divided into two
forms, (A) per se and (B) secundum accidens. These two forms are then respectively
broken down further, as (A) (i) genus into species, (ii) whole into parts, (iii) verbal
expression into significations, and as (B) (iv) subject into accidents, (v) accident into
subjects, (vi) accident into accidents.
24
Magee notes that the distinction of ‘earlier’ and ‘later’ Peripatetics could hardly
go back to Andronicus himself, if he means to refer to himself as a ‘later’ Peripatetic.
But certainly later sources refer to Andronicus as a ‘later’ Peripatetic (H b o æ
 æ ØŒ, Aspasius, in Eth. 44,18–45,5), and Andronicus need not have used the
word ‘later’, husteron, simply to distinguish himself from ‘earlier’ Peripatetics.
25
See Reinhardt (2007:518–21), Sharples (2008:281–2) and, on the Sophist passage
itself, Brown (1986).
 Andronicus’ Publication and Works 231
˛¯. ºº’ rÆ  ıªåøæE H Z ø a b ÆPa ŒÆŁ’ Æ , a b

æe ¼ººÆ Id º ª ŁÆØ . . . e  ª’ æ Id
æe æ ·
Visitor: But I think you’ll agree that among beings, some are said by
themselves (kath’ hauta), but some are always said in relation to other
things (pros alla) . . . but the different is always in relation to another . . .
Thus it is quite appropriate that Porphyry’s discussion of Andronicus’ favour
for division (and perhaps of Xenocrates’ as well) should have occurred in his
commentary on the Sophist. We also know that this passage of the Sophist
was read in antiquity (e.g. by Eudemus, ap. Simplic. in Phys. 99,25–31) as a
locus classicus for the distinction between substances and accidents. Por-
phyry, then, may have picked up Andronicus’ remarks and noted Androni-
cus’ view both in his Categories commentary and in his essay on division in
the Sophist. In further circumstantial support of this, we might point to a
passage of Alexander on the Metaphysics (in Met. 242,15–16 and 243,3),
where the bipartition of substance and accident is said to be the ‘first division’
found in the Categories (Bfi
æfiÅ ØÆØæ Ø H K ˚ÆŪæÆØ). I discuss
the preserved testimonies to Andronicus on this question in ch. 2.

Andronicus’ Paraphrase of the Categories

We also know of a paraphrase of the Categories, which was treated in more
depth in ch. 2 above. One reliable fact about Andronicus, upon which the
sources agree, was his focus upon the Categories as a foundational text of
Aristotelian studies, and his work to make it, and Aristotelian logic in general,
more comprehensible and systematic. Plotinus, judging from Simplicius 270,2
and 347,19, appears to have respected Andronicus’ treatment, and Simplicius
himself refers to it at length (cf. Plezia 1946: 6–10); it is not mentioned in
Simplicius’ full ‘honour roll’ (Barnes) at 1,3–2,29, where the limelight is won by
Boethus, and when Andronicus is mentioned with others he does not have first
place (63,21, 159,32, 202,1–5, 203,4). Simplicius calls this a paraphrasis
(26,17–20), a fairly generic term, and ‘Andronicus the paraphrasist’ contrasts
with ‘Boethus the exegete’ at 30,3–5. It is not clear whether this paraphrase
included the post-praedicamenta, which Andronicus is said to have discon-
nected from the rest of the work; but as I have suggested above, Andronicus
does not appear to have doubted that the post-praedicamenta were by Aristotle,
merely whether they belonged where he found them.
It may be possible that Andronicus made some of his critical or corrective
comments about the Categories in his paraphrasis—where he found Aris-
totle’s language ambiguous (Simplic. in Cat. 202,5) or insufficiently precise
(54,8–21).26 For example, he may have offered an alternative and remarked

26
See also the anonymous commentary in the Archimedes Palimpsest, 6,9–7,7 Chiar-
adonna et al. 2013, with commentary; Luna 2001b: 436–7; and Barnes 2003: 360–1.
232 Other Works of Interest
on the need for the alteration. However, the comments on the text that we
find in Simplicius could just as well come from a commentary that Simplicius
does not mention.

Andronican Spuria
Also ascribed to Andronicus are a first-century or second-century ce study
inscribed On Affections (—æd —ÆŁH , ed. Thirry, 1977) and a much later,
Renaissance text inscribed with his name. Thirry discusses the authorship
of the —æd —ÆŁH in detail, and as we shall see below, the tradition
demonstrates a very real interest in Andronicus’ ‘pathology’. Despite the
apparent inauthenticity of the —æd —ÆŁH , Andronicus does appear to have
offered a real definition of
ÆŁ that attracted the attention of later philo-
sophers, noted above (Aspasius, in Eth. 44–5), and made favourable remarks
about Xenocrates’ definition of the soul (ap. Themistium in de an. 32,23).
The Peri pathōn can be divided into two parts (cf. Gottschalk 1987, 1130
and Moraux I, 138 and following), the first a catalogue of passions and the
eupatheiai that the Stoic sage might feel, and the second a longer section of
virtues and vices. The first part seems thoroughly Stoic; the second a mixture
of Peripatetic and Stoic material. First in the second part comes the chapter
on the virtues, with lists of Stoic definitions; next a chapter entitled Kata
Chrusippon, giving Chrysippus’ definitions of phronesis and subordinate
virtues (though this is omitted in the oldest MSS and bracketed by Glibert-
Thirry: it states that aretē is a mean, giving a reference to EN 2 at c. 2–6).

Letters of Aristotle
Andronicus appears to have included a number of ‘Aristotelian’ letters in his
edition, including the famous letter of ‘Aristotle to Alexander’ (Aulus Gellius
20.5; cf. Plut. Alex. 7, 668bc, with Simplicius in Ph. 8,21–9), which he himself
uses to credit Aristotle with the publication of his esoteric works. Olympio-
dorus, writing in the sixth century ce (6,11–13) indicates that Andronicus
and Artemon ‘collected’ these letters; the text of Ptolemy the Unknown
(no. 96) suggests that Andronicus actually found these letters himself. In
any event, if he included them in his collection it follows that he (almost
certainly) viewed them as genuine witnesses. As these letters suppose the
publication of various Aristotelian school treatises during Aristotle’s own
lifetime, Andronicus might well have believed that such contemporary
publications really existed.
 APPENDIX 3

An Outline of Aristotle’s Categories

The structure, textual unity, authenticity, and function of the Categories, as

well as its place in Aristotle’s philosophy, have been considered in much
greater detail elsewhere.1 My intention here is just to survey the organization
of the treatise as it presented itself to the readers discussed throughout this
book, beginning with Andronicus of Rhodes.
1. The Ante-praedicamenta. The opening section, traditionally labelled
the ‘ante-praedicamenta’ (literally ‘what comes before the categories’),
comprises the first three traditional chapter headings (corresponding
to Bekker numbers 1a1–15, 1a16–1b9, and 1b10–24 respectively).
a. Chapter 1: The ‘-onymies’. Ch. 1 (1a1–15), sometimes called the
‘-onymies’ in modern scholarship, introduces three kinds of items.
These items are labelled, respectively, ‘homonyms’, ‘synonyms’, and
‘paronyms’. Roughly speaking, this chapter explores how, in ordin-
ary Greek, one term successfully picks out multiple referents, either
equivocally (homonymy), univocally (synonymy), or with a change
in the grammatical accidence of the term (paronymy). Things are
then carved up from the perspective of these ordinary linguistic
phenomena: for instance, a human being and an ox can both be
called ‘synonyms’ because, in ordinary Greek, they are both cor-
rectly addressed by the term ‘ÇfiH’ (animal).
The chapter proceeds as follows:
i. ‘Homonyms’ are items that are related to each other only in that
they can be addressed by the same referring term (hÆ) in
ordinary Greek. The examples offered are a human being and a
painting, both of which can be validly referred to by the noun
‘ÇfiH’ (which in ancient Greek is equivocal between ‘animal’ or
‘painting’). Besides the fact that I can validly address that paint-
ing and that person using the word ‘ÇfiH’, the painting and
person have nothing in common.
ii. ‘Synonyms’ are items that are related to each other in two ways:
like homonyms, they can be addressed by the same term in
Greek, but they are also related in that they share something

1
Frede 1987: 11–48, Bodéüs 2001: xi–clxxxviii, Mann 2000, Wedin 2000, and
Menn 1995; see also Ackrill 1963 for English translation and commentary.
234 An Outline of Aristotle’s Categories
called a ‘ºª B P Æ’ in common. It has been contentious
since antiquity just what a ‘ºª B P Æ’ is, and English
translations vary, but ‘statement of being, i.e. definition’ may not
be far off. The examples offered are a human being and an ox.
They may both be addressed by the term ‘ÇfiH’, but they also
share a definition in common, namely the definition answering
to the term ‘ÇfiH’: both persons and oxen are animals, i.e.
animated beings.
iii. ‘Paronyms’ are items that can be addressed by a name derived
from something else, with a difference of ending. For example, a
grammarian is so-called from grammar, and a brave person is
so-called from bravery.
It is today,2 and was in antiquity, hotly debated why the Cat. begins
with this chapter. Around the first century bce, for instance, An-
dronicus argued that some preliminary notion of homonyms, syn-
onyms, and paronyms was valuable for understanding the rest of
the treatise (Simplicius in Cat. 21,21–4), while various critics, such
as Lucius and Athenodorus, contended that the first chapter failed
to cohere with the rest of the work. An editor may have tacked it on
to the central portion of the treatise because he believed its explan-
ation of unfamiliar concepts or words would be useful for the
remainder. (That is more or less the explanation offered by the
commentators, although it is worth noting that the terminology of
the first chapter is not employed regularly throughout the rest of the
treatise.)
b. Chapter 2: ‘Things that are spoken’ and ‘things that exist’. The
second chapter heading deals with two groups of things in sequence:
i. Things that are said (which may be said with combination or
without combination). First (1a16–19), Aristotle considers ‘things
said’ (º ª Æ). These are divided (perhaps exhaustively) into two:

2
Michael Wedin (2000: 11–37) cites this as a question posed by early scholars:
‘The book begins without introductory remarks that give its purpose and object’
(Brandis 1833: 267), while this peculiar introduction makes for ‘the lack of any very
definite information as to Aristotle’s precise object in formulating’ the theory of
categories (Ross 1924, 1, lxxxii). It is also posed by Simplicius. One answer is found
in Ackrill (1963: 69), who follows the ancient Greek and Latin commentary tradition
in describing 1a1–15 as containing ‘certain preliminary points and explanations’
presupposed by the main body of the work. John Rist (1989a: 94), by contrast, lays
aside 1a1–15 as ‘loosely connected’ with the rest of the text. According to other
interpretations, as Barrington Jones rhetorically suggests (1972: 117), the first chapter
appears as ‘an incidental excrescence on the work’. Yet many critics, such as Jones
(1972), Dancy (1975), Frede (1987), Furth (1988), Lewis (1991), Wedin (2000), and
Mann (2000), hold that a coherent account of 1a1–15 is valuable for the interpretation
of the text.
 An Outline of Aristotle’s Categories 235
there are things said ‘without combination’, i.e. single terms con-
sidered on their own, such as ‘person’ and ‘runs’; and there are
things said ‘with combination’, i.e. complete sentences that include
both a noun and a verb, such as ‘person runs’. This division is
echoed in the De Interpretatione, where it is explored in much
greater detail.3 It is very important to Andronicus and the later
philosophers discussed in this book, who consider the former class
(‘things said without combination, i.e. simple terms signifying
simple items) to be the scope of the Categories, and anticipate the
Neoplatonic argument that for this reason the Categories describes
the building blocks of Aristotelian logic—since Aristotle proceeds
in De Int. to construct from these building blocks ‘combined’
sentences (signifying states of affairs), then syllogisms (Prior Ana-
lytics), and finally demonstrations (Posterior Analytics). (For a
thoughtful criticism of these arguments, see Morison 2005; for
their historical origins, see Bodéüs 2001: xiv–xvi, and the introduc-
tion to Hadot 1990.)
ii. Things that exist (which may be said-of but not in a subject, or
in a subject but not said-of a subject, or both said-of and in a
subject, or neither said-of nor in a subject). In this part of ch. 2
(1a20–b9), Aristotle seems to be concerned with real beings (ZÆ),
chiefly those whose names4 can validly stand in the predicate
position of a sentence like ‘Socrates is human.’ He is interested in
sentences that include a subject term (‘Socrates’), a copula (the
verb ‘is’), and a predicate term (‘human’): schematically, ‘A is B.’ (It
seems clear that Aristotle means to talk about predicative sen-
tences, although the technical verb ŒÆŪæBÆØ does not appear

3
Just what Aristotle means by ‘things spoken’ here is also a matter of heated
discussion: they may be the objects or states of affairs signified by linguistic utterances,
or the utterances themselves. The latter would be a reasonable guess, since the
‘combination’ described seems to apply more intuitively to words (the noun ‘person’
and verb ‘runs’ are ‘combined’ to make a sentence) than to things.
4
I try to distinguish clearly here between use and mention. As for the modern
complaint that Aristotle disregards the use–mention distinction, one might agree with
Bäck (2000) when he remarks that Aristotle appears to ‘switch back and forth from
speaking about words to speaking about real things’ and does so ‘quickly and blithely’
with confusing rapidity (133–4). Two recent approaches to this ambiguity are evident
in Wedin (2000) and De Rijk (2002:I). As a policy, Wedin will ‘follow Aristotle’s
practice of disregarding use–mention boundaries in formulating a number of his
theses’ (12 n. 6). By contrast, De Rijk, who expresses this equivocation as ‘the absence
of a clear-cut borderline between a linguistic expression [ . . . ] taken as a linguistic
tool, and its significate’, proposes that Aristotle exploits it intentionally and system-
atically, and christens this ‘the rule of indiscriminate reference’ (63–4)—a necessary
law and ‘main rule’ for his exegetical approach to Aristotelian semantics.
236 An Outline of Aristotle’s Categories
until 1b10.) He divides beings whose names can stand in the
predicate position of such a sentence into four kinds. This fourfold
division is constructed from two relationships that these beings
might hold towards the being whose name stands as the subject
( Π ) of the sentence.
1. The two relationships: said-of and in
a. The first relationship is called being said-of5 something as
a subject (ŒÆŁ’ Œ Ø ı Øe º ª ÆØ). It indicates an
ontological relation that is typically, but not invariably,
reflected in everyday speech. In general, B is said-of A as a
subject just in case we validly talk about ‘A’ as ‘a B’ (in Greek,
about ‘A’ as ‘B Ø’). Thus Socrates is a human, human is an
animal, and grammar is a (branch of) knowledge.6 (Interest-
ingly, Aristotle takes the fact that one can validly say ‘A is a B’
in ordinary Greek as a good indicator, at least heuristically,
that A really is a B.) In the framework of ten ŒÆŪæ ÆØ that
Aristotle will later outline, the said-of relationship operates
intra-categorically, or so to speak ‘vertically’: it relates the
universals within a category (those items which are ‘said of
many things’, De Int. 17a39–40)7 to the items of which they
are said in the same category. For instance, within the cat-
egory of reality (P Æ, traditionally rendered ‘substance’),8
genera like ‘animal’ are said-of species like ‘human’, while
species are said-of individual members of the species, which
are primary realities like Socrates (ch. 5, 2a11–18); likewise, in
the category of quality ( EÆ), knowledge is said-of a piece
of grammatical knowledge. Andronicus of Rhodes (ap. Sim-
plicium in Cat. 54,8–21) noticed that Aristotle’s linguistic
heuristic fails to guarantee this intra-categorial relation in
some cases: for example, we can validly say that ‘Socrates is

5
I am here following Lewis (1991) and Mann (2000) in capitalizing these names
to flag their technical usage by Aristotle.
6
For example, Aristotle states that human (¼Łæø ) is said-of an individual
human being (F Øe IŁæ ı), like Socrates, as a subject, and knowledge
(K ØÅ) is said-of an individual branch of knowledge, like grammatical expertise
(B ªæÆÆØŒB <K ØÅ>). More examples follow in ch. 3 (1b10–15), now using
the technical verb ‘predicate’: ‘human is predicated of the individual human
(¼Łæø  ŒÆa F Øe IŁæ ı ŒÆŪæ EÆØ), and animal of human . . . ’
7
Items that can be ‘said of many things’, in the language of De Int. 17a39–40.
8
I will sometimes avoid the traditional Latinate label ‘substance’, although it is
firmly entrenched, just because it is rather misleading in English; see Loux 1991:
15–16. Best of all is to leave ousia untranslated, but that can also be off-putting to the
Greekless reader. As Loux points out, ‘reality’ is too much like ‘being’ to clearly draw
out Aristotle’s distinction between ousia and on, but it has the advantage of being an
abstract noun in English, and it is reasonably clear.
 An Outline of Aristotle’s Categories 237
an Athenian’, implying that Athenian is said-of Socrates—
but ‘Athenian’ seems to pick out something in the category of
quality, while ‘Socrates’ picks out a reality.
b. The second relation is called being in a subject (K
Œ Ø ø fi ), which Aristotle describes as being ‘in some-
thing, not as a part, and unable to exist apart from what
it is in’. For instance, some grammatical knowledge is in
my mind, and pallor is in Socrates’ body (assuming his
body is pale). Aristotle’s description stresses that the
pallor is not constitutive of Socrates’ body—the body is
not a whole of which this colour is a constitutive part—
and likewise, my knowledge of Greek (say) is not consti-
tutive of my mind. This relation is trans-categorial: it
applies in cases where some quality, quantity, or another
of the latter nine categories ‘inheres’ in a reality (P Æ)
such as Socrates or Bucephalus. There is a linguistic
check that we can deploy to confirm this, since inherence
of B in A applies when I can say that ‘A is B’ but I cannot
say that ‘A is a B’: Socrates is not a pale (but pallor is in
him), and my mind is not a knowledge (but knowledge
is in it).
This distinction between two kinds of predication was taken to
be crucial in antiquity, especially as interest in Aristotelian
logic began to collect momentum after the first century bce.
Someone who accepts this subdivision of predication into the
said-of and in relations, based on the linguistic heuristics that
Aristotle highlights, might then be more willing to distinguish
the ontological inherence of a non-essential feature in a subject
(such as pallor in Socrates) from a subject’s essentially belong-
ing to a kind (such as Socrates’ belonging to the human
species). In this way, these relations also pave the way for
essentialism, which is critical for subsequent Aristotelian logic
and metaphysics. Ancient Peripatetics like Alexander of Aph-
rodisias criticized the Stoics for failing to appreciate just this
distinction in their logic, and treating all attributes as equal
(Alex. in Top. 42,27–43,2: see LS 32E). I argue in ch. 2 that
Andronicus placed considerable weight on the lessons of this
chapter, and was partly motivated by the perceived failure of
238 An Outline of Aristotle’s Categories
his predecessors to take proper stock of the difference between
essential and accidental predication.
2. The fourfold division

Cat. 2, 1a20–b9 Beings IN a subject Beings not IN a subject

Beings SAID-OF a Genera and species of items in
subject non-P Æ categories
Beings not SAID-OF Individual items in non-P Æ
a subject categories, such as  Ø
ªæÆÆØŒ (‘some particular
knowledge of grammar’)
Genera and species of primary
P ÆØ (i.e. secondary P ÆØ)
Primary P ÆØ, individuals,
e.g. › Ø ¼Łæø  (‘some
particular human being’)

a. Aristotle next uses these relations to produce a grid9 of four

kinds of beings, illustrated in the following table (adapted
from Reinhardt 2007: 515).
i. Beings that are IN a subject will prove to belong to the
non-P Æ categories (quantity, quality, relation, and so
on), and may be universals (if they are SAID-OF a
subject) or particulars (if they are not SAID-OF a
subject).
ii. Beings that are not IN a subject are realities (P ÆØ), and
may also be universals (if they are SAID-OF a subject) or
particulars (if they are not SAID-OF any subject). This
latter class—beings that are neither IN nor SAID-OF a
subject—are called ‘primary P ÆØ’: these are particular
beings such as Socrates and Bucephalus.
One of the most influential implications of this fourfold div-
ision in antiquity was the statement that particular beings, such
as Socrates and Bucephalus, count as ‘primary realities’. Some
Platonist critics of the Categories discussed in this book, such as
Lucius and Nicostratus (ch. 4), were alarmed, partly on this
basis, that the Categories omitted the intelligible Platonic
Forms as primary beings altogether. Aristotelian readers, like
Boethus of Sidon (ch. 5), introduced the question how this
account of primary P Æ as ‘neither SAID-OF nor IN a sub-
ject’ could jibe with the hylomorphic analysis of P Æ in terms

9
This may not literally have been intended as a diagram, although it is possible: for
Aristotle’s use of diagrams and tables in teaching, see recently Natali 2013: 114–15.
 An Outline of Aristotle’s Categories 239
of Form and Matter in Aristotle’s Metaphysics (Simplicius in
Cat. 78,4–20).
2. The Praedicamenta.
a. Initial list. With Cat. 4 (1b25–2a4), Aristotle introduces a list of ten
items ‘said without any combination’, which signify (ÅÆ  Ø) either
(1) realities (P ÆØ), (2) quantities, (3) qualities, (4) relations, (5)
where, (6) when, (7) position, (8) having, (9) acting, or (10) being
acted-on, and provides examples of each. We seem to have here not
just ten, but twenty, items. First, there are ten kinds of simple, referring
terms (‘things said without combination’, which, as Aristotle stresses
from 2a4–10, cannot produce an affirmation or denial). Secondly,
there are ten kinds of beings to which these terms refer. This latter
list of ten falls under the fourfold division just offered above: the P ÆØ
are those predicates that are not IN any subject; the remaining non-
P Æ items are those predicates that are IN subjects. (Within each
categorial ‘silo’ there will also be universal items SAID-OF more
specific items in the category.) The use of the verb ÅÆ  Ø, coupled
with the preceding divisions of ‘things said’ and ‘things that are’,
suggests that we have to do here with a relationship between simple
words meaning things. This implication was especially important in
antiquity, as the entire treatise Categories was taken to be concerned
with how simple words signify or refer to simple things—that is, to be
primarily concerned with semantics (cf. Simplicius in Cat. 13,11–18), or
with referring terms just insofar as they refer.
b. Detailed descriptions of the distinguishing characteristics of the
categories. These collected reflections on what is distinctive of each
category do not seem to constitute a proper definition of any one of
them, in terms of genus and differentiae, but then it would seem to be
impossible to ‘define’ an ultimate genus, which has no differentiae. The
ancient critics with whom this book is concerned, such as Boethus of
Sidon (ch. 6), also seem to have found in these descriptions a valuable,
non-technical ‘sketch account’ that could serve to ‘actuate’ beginners’
preconceptions of the highest genera, beginning the road to logic.10 As
Menn (1995) suggests, these brief, non-technical discussions of the
most distinctive features of each genus are certainly helpful in a

10
I refer several times in this book to Simplicius in Cat. 159,10–15, where Boethus
argues that it was not possible to give proper, formal definitions of the primary genera
at the early stage represented by Cat., but that Aristotle provides ªæÆçÆ or sketch
descriptions that can actuate or ‘stir up’ (IÆŒØ E) our concept (ØÆ) of a category
like the Relative. (That the central idea is Boethus’, I think, is strongly suggested by
this passage and others like 163,28–9; I also think that the verb IÆŒØ E is Boethus’
own, although it could also be a paraphrase by Simplicius or his source; it has strong
undertones in the Platonic tradition.)
240 An Outline of Aristotle’s Categories
pragmatic way, providing a set of rules of thumb for a dialectician to
recognize the genus to which a given term belongs.
i. Chapter 5: oPs a. Aristotle states that primary P ÆØ are distin-
guished by being neither SAID-OF nor IN a subject (2a11–18,
etc.)—which sparked some of the Platonist criticisms mentioned
above, by readers such as Lucius and Nicostratus. Species and
genera of P ÆØ are real, but secondary to the primary realities
themselves. Aristotle goes on to describe several other distinguish-
ing features or propria of P ÆØ: for instance, Aristotle notes that
everything that carries the name of an P Æ is so named synonym-
ously (3a33–b9, picking up the vocabulary of ch. 1); that every
P Æ seems to signify a certain ‘this’ (3b10, suggesting that again
we might have to do here with both a word, such as ‘human’, and a
being, this Socrates right here, and the semantic relationship
between them); that nothing is contrary to P ÆØ (3b24–32);
(iv) that they do not admit of a more or less (3b33–4a9); and
most of all, that what is numerically one and the same is able to
receive contraries (4a10–21).
ii. Chapter 6: quantity. Aristotle divides quantities into the discrete
and continuous, and those composed of parts with position and
those that are not so composed; he also gives a number of examples.
Among the distinguishing features of quantity that he mentions, he
notes that quantity has no contrary (5b11) and does not admit of
more and less (6a19). There were some criticisms of Aristotle’s
account of this category in antiquity, especially the decision to
include time and place under quantity; but they did not have the
broad range of the more fundamental questions about substance,
essentialism, and the general purpose of the book, noted above.
iii. Chapter 7: relative. Aristotle offers a short, crisp description of a
relative at 6a36–7: ‘We call relatives all such things as are said to be
just what they are, of or than other things, or in some other way in
relation to something else’ (tr. Ackrill). He gives examples such as
the ‘larger’, which is larger than something, and offers another
selection of distinguishing features of the category: relatives have
contrariety (6b15); they admit of more and less (6b19); they must
reciprocate (if slaves are of masters, then masters are of slaves, 6b28
and following); they are simultaneous by nature (7b15). But Aris-
totle raises a problem about whether the description given at the
start of the chapter can show that relatives are never spoken of as
substances, and this prompts him to offer what looks like a second
and different description of the relative at 8a28–34, as items ‘for
which being is the same as being related to something’. This second
description prompted all sorts of trouble in antiquity; Andronicus
 An Outline of Aristotle’s Categories 241
and his contemporary Ariston, for example, proposed that it was
circular (because the definition contained the definiendum), and
advanced an alternative definition of the relative as that whose
being was somehow related to something different (see ch. 2).11
iv. Chapter 8: quality. Aristotle divides qualities into states and con-
ditions, capacities and incapacities, affective qualities and affections
(which seem to have interested Andronicus, although this is largely
beyond the scope of the present study), shape, and (as a borderline
case) texture. He uses the terminology of paronymy from ch. 1 to
explain the relationship between a quality (like pallor) and the
qualified (the pale man), and offers an interesting reflection on
cases where ordinary language fails to reflect the ontological reality
(10a27–b11). Some of the features of qualification include the
following: it admits contrariety (10b12) and more and less
(10b26), and most distinctively, only in virtue of qualities ‘are things
called similar and dissimilar’ (11a15–19).
v. Chapter 9: Acting and Being Acted On. Very briefly (11b1–9),
Aristotle offers an account of two important features of these two
categories: they admit of contrariety and more and less. The rest of
the chapter (11b10–14)—which explains why there is no further
account of the categories’ position, when, where, and having—is an
interpolation.
3. The post-praedicamenta (‘what comes after the categories’). After a short
interpolated introduction, which simply states that ‘something must be
said about opposites . . . ’ (11b15–16), a new discussion begins, treating
opposition, priority and simultaneity, kinds of movement or change, and
the usage of the word ‘have’. Andronicus of Rhodes already maintained
that this section was appended ‘against the purpose of the treatise’
(ap. Boethium in Cat. 263B).

11
Other ancient readers began the tradition that the first ‘definition’ of the relative
might apply secundum dici, cued by the introductory phrase ‘We call . . . ’, while the
later ‘definition’ might apply secundum esse.
 Bibliography
Accattino, P. (1985). Alessandro di Afrodisia e Aristotele di Mitelene.
Elenchos 6, 67–74.
Ackrill, J. L. (1962). Critical Notice: Die Aristotelische Syllogistik. By Gün-
ther Patzig. Mind, 71, 107–17.
Ackrill, J. L. (1963). Aristotle: Categories and De Interpretatione. Oxford:
Oxford University Press.
Ackrill, J. L. (1997). Essays on Plato and Aristotle. Oxford: Oxford University
Press.
Adamson, P. (2005). On Knowledge of Particulars. Proceedings of the Aris-
totelian Society, 105, 257–78.
Adamson, P. (2007a). Knowledge of Universals and Particulars in the Bagh-
dad School. Documenti e studi sulla tradizione filosofica medievale, 18,
141–64.
Adamson, P. (2007b). Al-Kindī. New York: Oxford University Press.
Adamson, P., Baltussen, H., and Stone, M. W. F. (eds). (2004). Philosophy,
Science and Exegesis in Greek, Arabic, and Latin Commentaries. London:
Institute of Classical Studies, University of London.
Adamson, P., and Taylor, R. C. (eds). (2005). The Cambridge Companion to
Arabic Philosophy. Cambridge: Cambridge University Press.
Algra, K. A., van der Horst, P. W., and Runia, D. T. (eds). (1996). Polyhistor:
Studies in the History and Historiography of Ancient Philosophy Presented
to Jaap Mansfeld on his Sixtieth Birthday. Leiden: Brill.
Allen, J. (2005). The Stoics on the Origin of Language and the Foundations of
Etymology. In D. Frede and B. Inwood (eds), Language and Learning:
Philosophy of Language in the Hellenistic Age, pp. 14–35. Cambridge:
Cambridge University Press.
Allen, R. E. (1965). Studies in Plato’s Metaphysics. London: Routledge &
Kegan Paul.
Allen, R. E. (1969). Individual Properties in Aristotle’s Categories. Phronesis,
14, 31–9.
Aly, W. (1932). Strabo [3]. Realencyklopädie der classischen Altertumswis-
senschaft 4. 76–155.
Annas, J. (1974). Individuals in Aristotle’s Categories: Two Queries. Phron-
esis, 19, 146–52.
Anton, J. P. (1968). The Aristotelian Doctrine of Homonyma in the Cat-
egories and its Platonic Antecedents. Journal of the History of Philosophy,
6, 315–26.
Anton, J. P. (1969). Ancient Interpretations of Aristotle’s Doctrine of
‘Homonyma’. Journal of the History of Philosophy, 7, 1–18.
244 Bibliography
Armstrong, A. H. (1967). The Cambridge History of Later Greek and Early
Medieval Philosophy. Cambridge: Cambridge University Press.
Asmis, E. (1984). Epicurus’ Scientific Method. New York: Cornell University
Press.
Atherton, C. (1993). The Stoics on Ambiguity. Cambridge: Cambridge Uni-
versity Press.
Atherton, C. (2005). Lucretius on what language is not. In D. Frede and
B. Inwood (eds), Language and Learning: Philosophy of Language in the
Hellenistic Age, pp. 101–38. Cambridge: Cambridge University Press.
Aubenque, P. (ed.). (1980). Concepts et Catégories dans la pensée antique.
Paris: Vrin.
Austin, J. L. (1952). Critical Notice: Aristotle’s Syllogistic. By Łukasiewicz, J.’.
Mind, 61, 395–404.
Bäck, A. (2000). Aristotle’s Theory of Predication. Leiden: Brill.
Badawi, A. (1987). La Transmission de la philosphie grecque au monde arabe.
(2nd ed.). Paris: Vrin.
Baltussen, H. (2002). Philology or philosophy? Simplicius on the use of
quotations. In Worthington, I., and Foley, J. M. (eds) Epea and Gram-
mata: Oral and Written Communication in Ancient Greece, pp. 173–89.
Leiden: Brill.
Baltussen, H. (2007). From Polemic to Exegesis: The Ancient Philosophical
Commentary. Poetics Today, 28, 247–81.
Baltussen, H. (2008a). Philosophy and Exegesis in Simplicius: The Method-
ology of a Commentator. London: Duckworth.
Baltussen, H. (2008b). Dialectic in Dialogue: The Message of Plato’s Prota-
goras and Aristotle’s Topics. In A. McKay (ed.) Orality, Literacy, Memory
in the Ancient Greco-Roman World, pp. 203–26. Leiden: Brill.
Baltussen, H., Share, M., Atkinson, M., and Mueller, I. (2012). Simplicius On
Aristotle Physics 1.5-9. London: Bloomsbury Academic.
Barnes, J. (1971). Homonymy in Aristotle and Speusippus. The Classical
Quarterly, 21, 65–80.
Barnes, J. (1981). Boethius and the Study of Logic. In M. Gibson (ed.).
Boethius: His Life, Thought and Influence, pp. 73–89. Oxford: Basil
Blackwell.
Barnes, J. (1986). Peripatetic Negations. Oxford Studies in Ancient Philoso-
phy, 4, 201–14.
Barnes, J. (1993). Meaning, Saying, and Thinking. In K. Döring and T. Ebert
(eds) Dialektiker und Stoiker: Zur Logik der Stoa und ihrer Vorläufer
pp. 47–61, Stuttgart: F. Steiner.
Barnes, J. (1994). Aristotle: Posterior Analytics. Oxford: Clarendon Press.
Barnes, J. (1996). Grammar on Aristotle’s Terms. In M. Frede and G. Striker
(eds), Rationality in Greek Thought, pp. 175–202. Oxford: Oxford Univer-
sity Press.
 Bibliography 245
Barnes, J. (1997). Roman Aristotle. In J. Barnes and M. Griffin (eds),
Philosophia Togata II: Plato and Aristotle at Rome, pp. 1–70. Oxford:
Clarendon Press.
Barnes, J. (1999). Aristotle and Stoic Logic. In K. Ierodiakonou (ed.), Topics
in Stoic Philosophy, pp. 23–53. Oxford: Oxford University Press.
Barnes, J. (2003). Porphyry: Introduction. Oxford: Clarendon Press.
Barnes, J. (2005a). Les catégories et les Catégories. In O. Bruun and L. Corti
(eds), Les Catégories et leur histoire, pp. 11–80. Paris: Vrin.
Barnes, J. (2005b). What is a Disjunction? In D. Frede and B. Inwood (eds),
Language and Learning: Philosophy of Language in the Hellenistic Age,
pp. 274–98. Cambridge: Cambridge University Press.
Barnes, J. (2007). Peripatetic Logic. In R. Sorabji and R. W. Sharples (eds),
Greek and Roman Philosophy, 100 BC–200 AD (Vol. 2), pp. 531–46.
London: Institute of Classical Studies, University of London.
Barnes, J, Bobzien, S., Flannery, K. L., and Ierodiakonou, K. (1991). Alexan-
der of Aphrodisias on Aristotle Prior Analytics 1.1-7. London and Ithaca,
NY: Duckworth/Cornell University Publishing.
Barnes, J, and Griffin, M. (1997). Philosophia Togata II: Plato and Aristotle at
Rome. Oxford: Oxford University Press.
Barnes, J, Schofield, M., and Sorabji, R. K. (eds). (1979). Articles on Aristotle:
Metaphysics (Vol. 3). New York/London: St. Martin’s Press/Duckworth.
Barney, R. (2009). Simplicius: Commentary, Harmony and Authority. Anti-
quorum Philosophia, 3, 101–20.
Benson, H. H. (1988). Universals as Sortals in the Categories. Pacific Philso-
phical Quarterly, 69, 282–306.
Benson, H. H. (1989). A Note on Eristic and the Socratic Elenchus. Journal of
the History of Philosophy, 27, 591–99.
Bertier, J. (ed.). (1972). Mnésithée et Dieuchès. Leiden: Brill.
Biesterfeldt, H. H. (1973). Dass die Kräfte der Seele den Mischungen des
Körpers folgen. Stuttgart: Steiner.
Blank, D. (2007). The Life of Antiochus of Ascalon in Philodemus’ History of
the Academy and a Tale of Two Letters. Zeitschrift für Papyrologie und
Epigraphik, 162, 87–93.
Blumenthal, H. J. (1977). Review of P.L. Donini, Tre studi sull’ Aristotelismo
nel II secolo d.C. (Turin, 1974). Journal of Hellenic Studies, 97, 195.
Blumenthal, H. J. (1987). Alexander of Aphrodisias in the Later Greek
Commentaries on Aristotle’s De Anima. In J. Wiesner (ed.), Aristoteles.
Werk und Wirkung (Vol 2), pp. 90–103. Berlin and New York: Walter de
Gruyter.
Blumenthal, H. J. (1993). Soul and Intellect: Studies in Plotinus and Later
Neoplatonism. Aldershot, Hampshire; Brookfield, Vt.: Variorum.
246 Bibliography
Blumenthal, H. J., and Robinson, H. (eds). (1991). Oxford Studies in Ancient
Philosophy: Aristotle and the Later Tradition (Vol. Supplement). Oxford:
Clarendon Press.
Bobzien, S. (1996). Stoic Syllogistic. Oxford Studies in Ancient Philosophy, 14,
133–92.
Bobzien, S. (2005). The Stoics on fallacies of equivocation. In D. Frede and
B. Inwood (eds), Language and Learning: Philosophy of Language in the
Hellenistic Age, pp. 239–73. Cambridge: Cambridge University Press.
Bochenski, I. M. (1947). La Logique de Théophraste. Fribourg: La Librairie de
l’Université.
Bodéüs, R. (2001). Catégories—Aristote: texte établi et traduit par Richard
Bodéüs. Paris: Belles Lettres.
Bodéüs, R. (2008). Porphyre, Commentaire aux Catégories d’Aristote. Edition
critique Paris: Urin.
Boissevain, U. P. (1895). Cassii Dionis Cocceiani historiarum Romanarum
quae supersunt (Vol. 1). Berlin: Weidmann.
Bonazzi, M. (2007). Eudorus of Alexandria and Early Imperial Platonism. In
R. Sorabji and R. W. Sharples (eds), Greek and Roman Philosophy, 100
BC–200 AD (Vol. 2), pp. 365–77. London: Institute of Classical Studies,
University of London.
Bonazzi, M. (2008). The Commentary as Polemical Tool: The Anonymous
Commentator on the Theaetetus against the Stoics. Laval théologique et
philosophique, 64, 597–605.
Bonazzi, M. (2012). Antiochus and Platonism. In David Sedley (ed.), The
Philosophy of Antiochus, pp. 307–33. Cambridge: Cambridge University
Press.
Bonazzi, M. (2013). ‘Eudorus of Alexandria and the “Pythagorean” pseud-
epigrapha’. Pp. 385–404 in G. Cornelli, R. McKirahan, and C. Macris
(eds), On Pythagoreanism. Berlin: Walter De Gruyter.
Bonazzi, M., and Celluprica, V. (2005). L’eredità platonica. Studi sul plato-
nismo da Arcesilao a Proclo. Naples: Bibliopolis.
Bonazzi, M., and Opsomer, J (eds). (2009). The Origins of the Platonic
System: Platonisms of the Early Empire and Their Philosophical Contexts.
Collection d’études classiques 23. Louvain: Peeters–Société des études
classiques.
Bonitz, H. (ed.). (1870). Index Aristotelicus (Vol. 5). Berlin: Walter de
Gruyter.
Borgen, P., Fuglseth, K., and Skarsten, R. (2000). The Philo Index: A Complete
Greek Word Index to the Writings of Philo of Alexandria. Leiden: Brill.
Bostock, D. (1994). Aristotle: Metaphysics Books Z and ˙. Oxford: Claren-
don Press.
Boulluec, A. (1975). L’allégorie chez les Stoïciens. Poetique, 23, 301–22.
Boys-Stones, G. (2001). Post-Hellenistic Philosophy: A study of its develop-
ment from the Stoics to Origen. Oxford: Oxford University Press.
 Bibliography 247
Boys-Stones, G. (ed.). (2003a). Metaphor, Allegory, and the Classical Trad-
ition: Ancient Thought and Modern Revisions. Oxford: Oxford University
Press.
Boys-Stones, G. (2003b). The Stoics’ Two Types of Allegory. In G. Boys-
Stones (2003a), pp. 189–216. Oxford: Oxford University Press.
Boys-Stones, G. (2005). Alcinous, Didaskalikos 4: In Defence of Dogmatism.
In M. Bonazzi and V. Celluprica (eds), L’eredità platonica. Studi sul
platonismo da Arcesilao a Proclo, pp. 203–34. Naples: Bibliopolis.
Boys-Stones, G. (2007). Fallere Sollers: The Ethical Pedagogy of the Stoic
Cornutus. In R. Sorabji and R. W. Sharples (eds), Greek and Roman
Philosophy 100 BC–200 AD (Vol. 1), pp. 77–88. London: Institute of
Classical Studies, University of London.
Brennan, T. (2005). The Stoic Life: Emotions, Duties, and Fate. Oxford:
Clarendon Press.
Brink, K. O. (1940). Peripatos. Paulys Realencyclopädie, suppl. 7, cols.
899–949. Stuttgart: J. B. Metzler.
Brittain, C. (2001). Philo of Larissa: The Last of the Academic Sceptics.
Oxford: Oxford University Press.
Brittain, C. (2005) Common sense: concepts, definition and meaning in and
out of the Stoa. In D. Frede and B. Inwood (eds), Language and Learning:
Philosophy of Language in the Hellenistic Age, pp. 164–209. Cambridge:
Cambridge University Press.
Brower, J. (2001, revised 2014). Medieval Theories of Relations. The Stanford
Encyclopedia of Philosophy (Spring 2014 Edition) E. Zalta (ed.). <http://
plato.stanford.edu/entries/relations-medieval/>
Brown, L. (1986). Being in the Sophist: A Syntactic Inquiry. Oxford Studies in
Ancient Philosophy, 4 [Festschrift for J. L. Ackrill], 49–70. Repr. in G. Fine
(ed.). Plato 1: Metaphysics and Epistemology, pp. 455–78. Oxford: Oxford
University Press.
Brunschwig, J. (1991). On a Book-Title by Chrysippus: On the fact that ancients
admitted dialectic along with demonstrations. In H. J. Blumenthal and
H. Robinson (eds), Oxford Studies in Ancient Philosophy (Vol. Suppl.),
pp. 81–95. Oxford: Oxford University Press.
Brunschwig, J. (ed.). (1994a). Papers in Hellenistic Philosophy. Cambridge:
Cambridge University Press.
Brunschwig, J. (1994b). Epicurus and the Problem of Private Language. In
J. Brunschwig (1994a), pp. 21–38. Cambridge: Cambridge University
Press.
Brunschwig, J., and Nussbaum, M. C. (1993). Passions and Perceptions:
Studies in Hellenistic Philosophy of Mind. Cambridge: Cambridge Univer-
sity Press.
Bruun, O., and Corti, L. (2005). Les Catégories et leur histoire. Paris: Vrin.
248 Bibliography
Buffière, F. (1956). Les Mythes d’Homère et la pensée grecque. Paris: Les Belles
Lettres.
Burkert, W., Gemelli, L., Matelli, E., and Orelli, L. (1998). Fragmentsamm-
lungen philosophischer Texte der Antike—Le raccolte dei frammenti di
filosofi antichi (Vol. 3). Göttingen: Vandenhoeck & Ruprecht.
Burnyeat, M. F. (1982). Gods and Heaps. In M. Schofield and M.C. Nussbaum
(eds), Language and Logos, pp. 315–38. Cambridge: Cambridge University
Press.
Burnyeat, M. F. (1987). The Inaugural Address: Wittgenstein and Augustine
De Magistro. Proceedings of the Aristotelian Society, Supplementary
Volume 61, 1–24.
Burnyeat, M. F. (2005). Archytas and Optics. Science in Context, 18, 35–53.
Busse, A. (1887). Porphyrii isagoge et in Aristotelis categorias commentarium
(CAG 4.1). Berlin: Reimer.
Busse, A. (1888). Dexippi in Aristotelis categorias commentarium (CAG 4.2).
Berlin: Reimer.
Busse, A. (1895). Ammonii in Aristotelis categorias commentarius (CAG 4.4).
Berlin: Reimer.
Busse, A. (1898). Philoponi (olim Ammonii) in Aristotelis categorias com-
mentarium (CAG 13.1). Berlin: Reimer.
Busse, A. (1900). Eliae (olim Davidis) in Aristotelis categorias commentarium
(CAG 18.1). Berlin: Reimer.
Canick, H., and Schneider, H. (eds). (1996). Der Neue Pauly: Enzyklopädie
der Antike. Stuttgart: J.B. Metzler.
Canto-Sperber, M., and Pellegrin, P. (eds). (2002). Le Style de La Pensée:
Recueil de textes en hommage à Jacques Brunschwig. Paris: Les Belles
Lettres.
Capelle, W. (1927). Lukios [1]. Paulys Realencyclopdädie der classischen
Altertumswissenschaft, rev. Wissowa, vol. xii 2. Stuttgart: J. B. Metzler.
Caston, V. (1997). Epiphenomenalisms, Ancient and Modern. The Philo-
sophical Review, 106, 309–63.
Caston, V. (1998). Aristotle and the Problem of Intentionality. Philosophy
and Phenomenological Research, 58, 249–98.
Caston, V. (1999). Something and Nothing: The Stoics on Concepts and
Universals. Oxford Studies in Ancient Philosophy,17, 145–213.
Caston, V. (2008). Intentionality in Ancient Philosophy. The Stanford
Encyclopedia of Philosophy (Fall 2008 Edition) E. Zalta (ed.). <http://
plato.stanford.edu/entries/intentionality-ancient/>
Cauville, S. (1997). Le temple de Dendara: Les chapelles osiriennes: Commen-
taire (Vol. 1). Institut français d’archéologie orientale du Caire.
Celluprica, V. (ed.). (2003). Il libro B della Metafisica di Aristotele. Atti del
Colloquio Roma 30 novembre–1 dicembre 2000. Naples: Bibliopolis.
 Bibliography 249
Celluprica, V., and D’Ancona, C. (eds). (2004). Aristotele e i suoi esegeti
neoplatonici. Logica e ontologia nelle interpretazioni greche e arabe. Atti del
Convegno internazionale Roma 19–20 ottobre 2001. Roma: Bibliopolis.
Chaniotis, A. (2004). Epigraphic evidence for the philosopher Alexander of
Aphrodisias. Bulletin of the Institute of Classical Studies, 47, 79–81.
Charles, D. (2000). Aristotle on Meaning and Essence. Oxford: Oxford Uni-
versity Press.
Charles, D. (2010). Definition in Greek Philosophy. Oxford: Oxford Univer-
sity Press.
Chase, M. (2000). Études sur le commentaire de Porphyre aux Catégories
d’Aristote adressé à Gédalios. Annuaire EPHE, Section des sciences reli-
gieuses 108, 505–10.
Chase, M. (2003). Simplicius: On Aristotle’s Categories 1–4. London:
Duckworth.
Chase, M. (2010). Porphyry on the Cognitive Process. Ancient Philosophy,
30, 383–405.
Chase, M. (2011). Individus et descriptions. Contribution à une histoire du
problème de la connaissance des individus dans la philosophie néoplato-
nicienne. Revue des Sciences philosophiques et théologiques, 95,1, 3–36.
Cherniss, H. (1944). Aristotle’s Criticism of Plato and the Academy. Balti-
more: Johns Hopkins Press.
Chiaradonna, R. (2002). Sostanza, Movimento, Analogia. Plotino critico di
Aristotele. Naples: Bibliopolis.
Chiaradonna, R. (2004). Plotino e la teoria degli universali. Enn. VI 3 [44], 9.
In V. Celluprica and C. D’Ancona (eds), Aristotele e i suoi esegeti neopla-
tonici. Logica e ontologia nelle interpretazioni greche e arabe, pp. 1–35.
Naples: Bibliopolis.
Chiaradonna, R. (2005). Plotino e la corrente antiaristotelica del platonismo
imperiale: analogie e differenze. In M. Bonazzi and V. Celluprica (eds),
L’eredità platonica. Studi sul platonismo da Arcesilao a Proclo, Elenchos
45, pp. 235–74. Naples: Bibliopolis.
Chiaradonna, R. (2007a). Porphyry’s Views on the Immanent Incorporeals.
In G. Karamanolis and A. Sheppard (eds), Studies on Porphyry (BICS
Suppl. 98), pp. 35–49. London: Institute of Classical Studies, University
of London.
Chiaradonna, R. (2007b). Porphyry and Iamblichus on Universals and
Synonymous Predication. Documenti e studi sulla tradizione filosofica
medievale, 18, 123–40.
Chiaradonna, R. (2007c). Platonismo e teoria della conoscenza stoica tra II e
III secolo D.C. In M. Bonazzi and C. Helmig (eds), Platonic Stoicism—
Stoic Platonism: The Dialogue Between Platonism and Stoicism in
Antiquity, pp. 209–42. Leuven: Leuven University Press.
250 Bibliography
Chiaradonna, R. (2008). What is Porphyry’s Isagoge? Documenti e studi sulla
tradizione filosofica medievale, 19, 1–30.
Chiaradonna, R. (2009). Autour d’Eudore: Les débuts de l’exégèse des Cat-
égories dans les Moyen Platonisme. In M. Bonazzi and J. Opsomer (eds),
The Origins of the Platonic System: Platonisms of the Early Empire and
their Philosophical Contexts, pp. 89–111. Louvain/Namur/Paris/Walpole,
Ma: Éditions Peeters—Société des études classiques.
Chiaradonna, R. (2011). Plotinus’ Account of the Cognitive Powers of the
Soul: Sense Perception and Discursive Thought. Topoi, 31, 191–207.
Chiaradonna, R. (2013). Platonist approaches to Aristotle: from Antiochus of
Ascalon to Eudorus of Alexandria (and beyond). In M. Schofield (ed.),
Aristotle, Plato, and Pythagoreanism in the First Century BC, pp. 28–52.
Cambridge: Cambridge University Press.
Chiaradonna, R., and Galluzzo, G.(eds). (2013). Universals in Ancient Phil-
osophy. Pisa: Edizioni della Normale.
Chiaradonna, R., Rashed, M., and Sedley, D. (with Natalie Tchernetska).
(2013). A Rediscovered Categories Commentary. Oxford Studies in
Ancient Philosophy, 44, 129–94.
Chiaradonna, R., and Trabattoni, F. T. (2009). Physics and Philosophy of
Nature in Greek Neoplatonism: Proceedings of the European Science Foun-
dation Exploratory Workshop (Il Ciocco, Castelvecchio Pascoli, June 22–24,
2006). Leiden: Brill.
Chlup, R. (2012). Proclus: An Introduction. Cambridge: Cambridge Univer-
sity Press.
Cochiarella, N. B. (2001). Logic and Ontology. Aristomathes, 12, 117–50.
Code, A. (1984). The Aporematic Approach to Primary Being in Metaphys-
ics Z. In J. Pelletier and J. King-Farlow (eds), Canadian Journal of Phil-
osophy, Supplement vol. 10, pp. 1–20. Guelph: Canadian Association for
Publishing in Philosophy.
Cohen, S.M., and Matthews, G.B. (1991). Ammonius On Aristotle Categories.
London: Duckworth.
Cooper, J. M. (2012). Pursuits of Wisdom: Six Ways of Life in Ancient
Philosophy from Socrates to Plotinus. Princeton: Princeton University
Press.
Corcoran, J. (1972). Completeness of an Ancient Logic. Journal of Symbolic
Logic, 37, 696–702.
Corcoran, J. (1974a). Ancient Logic and its Modern Interpretations. Dor-
drecht: D. Reidel.
Corcoran, J. (1974b). Aristotle’s Natural Deduction System. In J. Corcoran
(1974a), pp. 85–131. Dordrecht: D. Reidel.
Corcoran, J. (1994). The Founding of Logic: Modern Interpretations of
Aristotle’s Logic. Ancient Philosophy, 14 (Special Issue), 9–24.
 Bibliography 251
Cornelli, G., McKirahan, R., and Constantinos, M. (eds). (2013). On Pytha-
goreanism. Berlin: Walter de Gruyter.
Cornford, F. M. (1935). Plato’s Theory of Knowledge. London: Routledge &
Kegan Paul.
Courcelle, P. (1969). Late Latin Writers and their Greek Sources.
(H. E. Weddeck, trans.) (2nd ed.). Cambridge, MA: Harvard University
Press.
D’Ancona, C. (2002). Commenting on Aristotle: From Late Antiquity to the
Arab Aristotelianism. In Geerlings and Schulze (eds), Der Kommentar in
Antike und Mittelalter, pp. 201–51. Leiden: Brill.
Dancy, R. (1975). On Some of Aristotle’s First Thoughts about Substances.
The Philosophical Review, 84, 338–73.
De Haas, F. A. J. (2001). Did Plotinus and Porphyry Disagree on Aristotle’s
Categories? Phronesis, 46, 492–526.
De Haas, F. A. J., and D’Ancona, C. (2004). Context and strategy of Plotinus’
treatise on the genera of being (Enn. VI.1-3 [42–44]). In V. Celluprica and
C. D’Ancona (eds), Aristotele e i suoi esegeti neoplatonici. Logica e ontologia
nelle interpretazioni greche e arabe (eds), pp. 37–53. Naples: Bibliopolis.
De Haas, F. A. J., and Fleet, B. (2001). Simplicius on Aristotle: Categories 5–6.
London/Ithaca, NY: Duckworth/Cornell University Press.
De Rijk, L.M. (1895). Simplicii in Aristotelis physicorum libros octo commen-
taria (CAG 10). Berlin: Reimer.
De Rijk, L. M. (1951). The Authenticity of Aristotle’s Categories. Mnemos-
yne, 4, 129–59.
De Rijk, L.M. (1996). On Aristotle’s Semantics in De Interpretatione 1–4. In
K. A. Algra, P. W. van der Horst, and D. T. Runia (eds), Polyhistor: Studies
in the History and Historiography of Ancient Philosophy, pp. 115–34.
Leiden: Brill.
De Rijk, L.M. (2002). Aristotle: Semantics and Ontology (Vol. 1). Leiden: Brill.
DeLancey, C. (2002). Passionate Engines. Oxford: Oxford University Press.
Deslauriers, M. (1990). Plato and Aristotle on Division and Definition.
Ancient Philosophy, 10, 203–19.
Deslauriers, M. (2007). Aristotle on Definition. Leiden: Brill.
Devereux, D., and Pellegrin, P. (eds). (1990). Biologie, Logique, et Métaphy-
sique chez Aristote. Paris: CNRS.
Diels, H. (1879). Doxographi Graeci. Berlin: Reimer.
Diels, H. (1882). Simplicii in Aristotelis physicorum libros octo commentaria
(CAG 9). Berlin: Reimer.
Dillon, J. M. (1977, 2nd ed. 1996). The Middle Platonists: A Study of
Platonism 80 B.C. to A.D. 220. London: Duckworth.
Dillon, J. M. (1981). Antiochus and the Late Academy, by John Glucker. The
Classical Review, 31, 60–2.
252 Bibliography
Dillon, J. M. (1990). Dexippus on Aristotle’s Categories. London/Ithaca, NY:
Duckworth/Cornell University Press.
Dillon, J. M. (tr.). (1993). Alcinous. The Handbook of Platonism. Oxford:
Clarendon Press.
Dillon, J. M. (1997). Iamblichus’ Noera Theoria of Aristotle’s Categories.
Syllecta Classica, 8, 65–77.
Dillon, J. M. (1998). Gathering Fragments: The Case of Iamblichus. In
W. Burkert, L. Gemelli, and E. Matelli (eds), Fragment-sammlungen phi-
losophischer Texte der Antike—Le raccolte dei frammenti di filosofi antichi.
Aporemata. Kritische Studien zur Philologiegeschichte. Band 3, pp. 161–81.
Göttingen: Vandenhoeck & Ruprecht.
Dillon, J. M. (1999). A Case-Study in Commentary: The Neoplatonic Exe-
gesis of the Prooimia of Plato’s Dialogues. In G. Most (ed.), Commentar-
ies—Kommentare. Aporemata: Kritische Studien zur Philologiegeschichte,
Band 4, pp. 206–22. Göttingen: Vandenhoeck & Ruprecht.
Dillon, J. M. (2000). Eudore d’Alexandrie. In Richard Goulet (ed.), Diction-
naire des philosophes antiques. Band 3. Paris: CNRS, 2000, 290–3.
Dillon, J. M. (2003). The Heirs of Plato: A Study of the Old Academy (347–274
BC). Oxford: Oxford University Press.
Dillon, J.M., and Long, A. (eds). (1988). The Question of ‘Eclecticism’: Studies
in Later Greek Philosophy. Berkeley: University of California Press.
Dodds, E. R. (1951). The Greeks and the Irrational. Berkeley: University of
California Press.
Dodds, E. R. (1963). Proclus, The Elements of Theology: A Revised Text with
Translation, Introduction, and Commentary. Oxford: Oxford University
Press.
Donini, P. (1971). L’anima e gli elementi nel De Anima di Alessandro di
Afrodisia. Atti dell’ Accademia delle Scienze di Torino: classe di scienze
morali, storiche, e filologiche, 150, 61–107.
Donini, P.L. (1977). ‘L’aristotelismo nel primo secolo. La ricostruzione di
Paul Moraux’, Rivista di Filologia e di Istruzione Classica 105, 237–51.
Donini, P. (1982). Le scuole, l’anima, l’impero: la filosofia antica da Antioco a
Plotino. Torino: Rosenberg & Sellier.
Donini, P. (1988). The history of the concept of eclecticism. In J.M. Dillon
and A. Long (eds), The Question of Eclecticism: Studies in Later Greek
Philosophy (Hellenistic Culture and Society), pp. 15–33. Berkeley: Univer-
sity of California Press.
Döring, K., and Ebert, T. (eds). (1993). Dialektiker und Stoiker: Zur Logik der
Stoa und ihrer Vorläufer. Stuttgart: Franz Steiner Verlag.
Dörrie, H. (1944). Der Platoniker Eudoros von Alexandria, Hermes 79,
25–38.
 Bibliography 253
Dörrie, H. (1959). Porphyrios’ ‘Symmikta zetemata’: ihre Stellung in System
und Geschichte des Neuplatonismus, nebst einem Kommentar zu den
Fragmenten. Munich: C. H. Beck Verlag.
Dörrie, H. (1976). Platonica minora. Munich: W. Fink.
Dörrie, H., and Baltes, M. (eds). (1987). Der Platonismus in der Antike:
Grundlagen, System, Entwicklung. Stuttgart: Frommann-Holzboog Verlag.
Dörrie, H. and Baltes, M. (eds). (2002). Die philosophische Lehre des Plato-
nismus [3]: Der Platonismus in der Antike 6. Frommann-Holzboog Verlag.
Dupréel, E. (1909). ‘Aristot et le traite des catégories’, Archiv für Geschichte
der Philosophie 22, 230–51.
Düring, I. (1956). ‘Ariston or Hermippus? A Note on the Catalogue of
Aristotle’s Writings, Diog. L. v. 22’, Classica et Medievalia, 17, 11–21.
Düring, I. (1957). Aristotle in the Ancient Biographical Tradition. Göteborg:
Institute of Classical Studies of the University of Göteborg.
Dyson, H. (2009). Prolepsis and Ennoia in the Early Stoa. Berlin: Walter de
Gruyter.
Ebbesen, S. (1990). Porphyry’s legacy to logic: A reconstruction. In
R. Sorabji, Aristotle Transformed: The Ancient Commentators and Their
Influence, pp. 141–72. Ithaca, NY: Cornell University Press.
Ebbesen, S. (2001). Review of John Magee, Anicii Manlii Severini Boethii ‘De
divisione liber’. Speculum, 76, 1003–4.
Ebbesen, S. (2005). Theories of language in the Hellenistic age and in the
twelfth and thirteenth centuries. In D. Frede and B. Inwood (eds), Lan-
guage and Learning: Philosophy of Language in the Hellenistic Age,
pp. 299–319. Cambridge: Cambridge University Press.
Ellis, J. (1994). Alexander’s Defense of Aristotle’s Categories. Phronesis, 39,
69–89.
Evangeliou, C. (1988). Aristotle’s Categories and Porphyry. Leiden: Brill.
Evangeliou, C. (2006). Hellenic philosophy: origin and character. Surrey, UK:
Ashgate.
Falcon, A. (2008). The Pre-History of the Commentary Tradition: Aristote-
lianism in the First Century bce (Prolegomena to a Study of Xenarchus of
Seleucia). Laval théologique et philosophique:, 64, 7–18.
Falcon, A. (2011). Aristotelianism in the First Century bce: Xenarchus of
Seleucia. Cambridge: Cambridge University Press.
Falcon, A. (n.d.). Commentators on Aristotle. Retrieved 28 July 2012,
from <http://www.science.uva.nl/~seop/archives/fall2008/entries/aristotle-
commentators/>
Fazzo, S. (2002). Alexandre d’Aphrodise contre Galien: la naissance d’une
légende. Philosophie Antique: Problèmes, Renaissances, Usages—Questions
aristotéliciennes: Questions aristotéliciennes, 2, 109–44.
Fazzo, S. (2004). Aristotelianism as a Commentary Tradition. Bulletin of the
Institute of Classical Studies, 47, 1–19.
254 Bibliography
Ferejohn, M. T. (1980). Aristotle on Focal Meaning and the Unity of Science.
Phronesis, 25, 117–28.
Fine, G. (1993). On Ideas: Aristotle’s Criticism of Plato’s Theory of Forms.
Oxford: Oxford University Press.
Flannery, K. L. (1994). Ways into the Logic of Alexander of Aphrodisias.
Leiden: Brill.
Fleet, B. (2002). Simplicius on Aristotle: Categories 7–8. London/Ithaca, NY:
Duckworth/Cornell University Press.
Flew, A. (ed.). (1953). Logic and Language: Second Series. Oxford: Basil
Blackwell.
Fortenbaugh, W. W., and Mirhady, D. C. (1994). Peripatetic Rhetoric After
Aristotle. New Jersey: Transaction Publishers.
Fraser, P. M. (1972). Ptolemaic Alexandria. Oxford: Clarendon Press.
Frede, D. (2012). The Endoxon Mystique: What Endoxa Are and What They
Are Not. Oxford Studies in Ancient Philosophy, 43, 185–215.
Frede, D., and Inwood, B. (eds). (2005). Language and Learning: Philosophy
of Language in the Hellenistic Age. Cambridge: Cambridge University
Press.
Frede, M. (1987). Essays in Ancient Philosophy. Oxford: Clarendon Press.
Frede, M. (1989). Chaeremon der Stoiker. In W. Haase (ed.), Aufstieg und
Niedergang der römischen Welt (Vol. 36.3, pp. 2067–103). Berlin: Walter
de Gruyter.
Frede, M. (1990). The Definition of Sensible Substance in Metaphysics Z. In
D. Devereux and P. Pellegrin (eds), Biologie, Logique et Métaphysique chez
Aristote, pp. 113–29. Paris: CNRS.
Frede, M. (1999). Epilogue. In K. Algra, J. Barnes, J. Mansfeld, and
M. Schofield (eds), The Cambridge History of Hellenistic Philosophy,
pp. 771–97. Cambridge: Cambridge University Press.
Frede, M., and Patzig, G. (eds). (1988). Aristoteles, ‘Metaphysik Z’: Text,
Übersetzung und Kommentar (Vol. 2). Munich: C. H. Beck Verlag.
Frede, M., and Striker, G. (eds). (1996). Rationality in Greek Thought.
Oxford: Clarendon Press.
Frege, G. (1892). Uber Sinn und Bedeuntung. Zeitschrift für philosophie und
philosophische kritik, 100, 25–50.
Funk, K. (1907). Untersuchungen über die lucianische ‘Vita Demonactis.’
Philologus, Supplement (10), 558–674.
Furley, D. (ed.). (1999). From Aristotle to Augustine (Vol. 2). London:
Routledge.
Furley, D. J., and Nehamas, A. (1994). Aristotle’s Rhetoric: Philosophical
Essays. Princeton: Princeton University Press.
Furth, M. (1988). Substance, Form, and Psyche: An Aristotelean Metaphysics.
Cambridge: Cambridge University Press.
 Bibliography 255
Gaskin, R. (2000). Simplicius on Aristotle: Categories 9–15. London:
Duckworth.
Gaukroger, S. (2001). Francis Bacon and the Transformation of Early-
Modern Philosophy. Cambridge: Cambridge University Press.
Geerlings, W., and Schulze, C. (eds). (2002). Der Kommentar in Antike und
Mittelalter: Beitrage zu seiner Erforschung. (Vol. 1). Leiden/Boston/Köln:
Brill.
Gera, D. L. (2003). Ancient Greek Ideas on Speech, Language, and Civiliza-
tion. Oxford: Oxford University Press.
Gercke, A. (1891). ‘Ursprung der Aristotelischen Kategorien’, Archiv für
Geschichte der Philosophie 4, 424–41.
Gerson, L. P. (ed.). (1996). The Cambridge Companion to Plotinus. Cam-
bridge: Cambridge University Press.
Gerson, L. P. (2005). What is Platonism? Journal of the History of Philosophy,
43, 253–76.
Gerson, L. P. (ed.). (2011). The Cambridge History of Philosophy in Late
Antiquity. Cambridge: Cambridge University Press.
Gibson, M. (ed.). (1981). Boethius: His Life, Thought, and Influence. Oxford:
Blackwell.
Gibson, R. K., and Kraus, C. S. (eds). (2002). The Classical Commentary:
Histories, Practices, Theory. Leiden: Brill.
Gill, C. (2003). The School in the Roman Imperial Period. In B. Inwood (ed.),
The Cambridge Companion to the Stoics, pp. 33–58. Cambridge: Cam-
bridge University Press.
Gill, C. (2007). Galen and the Stoics: Mortal Enemies or Blood Brothers?
Phronesis, 52, 88–120.
Gill, M. L., and Pellegrin, P. (eds). (2008). A Companion to Ancient Philoso-
phy. New York: John Wiley & Sons.
Gillespie, C. M. (1925). The Aristotelian Categories. The Classical Quarterly,
19, 75–84.
A. Gioè (2002). Filosofi medioplatonici del II secolo d.c.: testimonianze e
frammenti. Gaio, Albino, Lucio, Nicostrato, Tauro, Severo, Arpocrazione.
Naples, 2002.
Gilbert-Thirry, A. (ed.). (1977). Pseudo-Andronicus de Rhodes’ ‘—æd
—ÆŁH’: Édition critique du texte grec et de la traduction latine médiévale
(Vol. Supplement). Leiden: Brill.
Gottschalk, H.B. (1972). ‘Notes on the Wills of the Peripatetic Scholarchs’,
Hermes, 100, 335–42.
Gottschalk, H. B. (1987). Aristotelian Philosophy in the Roman World from
the Time of Cicero to the End of the Second Century ad. In Wolfgang
Haase (ed.), Aufstieg und Niedergang der römischen Welt (Vol. 36.2,
pp. 1079–174). Berlin & New York: Walter de Gruyter.
256 Bibliography
Gottschalk, H. B. (1990). The Earliest Aristotelian Commentaries. In
R. Sorabji, Aristotle Transformed, pp. 55–81. Ithaca, NY: Cornell Univer-
sity Press.
Goulet, R. (1989). Dictionnaire des philosophes antiques. Tome I: d’Abam(m)
on à Axiothéa. Paris: CNRS.
Goulet, R. (1994). Dictionnaire des philosophes antiques. Tome II: de Babé-
lyca d’Argos à Dyscolius. Paris: CNRS.
Goulet, R. (2000). Dictionnaire des philosophes antiques. Tome III: d’Eccélos à
Juvénal. Paris: CNRS.
Goulet, R. (2003). Dictionnaire des philosophes antiques. Supplément. Paris:
CNRS.
Goulet, R. (2005). Dictionnaire des philosophes antiques. Tome IV: de Labeo à
Ovidius. Paris: CNRS.
Goulet, R. (2012a). Dictionnaire des philosophes antiques. Tome Va: de
Paccius à Plotin. Paris: CNRS.
Goulet, R. (2012b). Dictionnaire des philosophes antiques. Tome Vb: de
Plotina à Rutilus Rufus. Paris: CNRS.
Graeser, A. (1977). On Language, Thought, and Reality in Ancient Greek
Philosophy. Dialectica, 31, 359–88.
Graver, M. (1999). Philo of Alexandria and the Origin of the Stoic Pro-
patheiai. Phronesis, 44, 300–25.
Griffin, M., and Barnes, J. (eds). (1989). Philosophia Togata: Essays on
Philosophy and Roman Society (Vol. 1). Oxford: Oxford University Press.
Griffin, M. J. (2009). The Reception of Aristotle’s Categories, c.80 bc–ad 220.
University of Oxford: D.Phil. <http://ora.ox.ac.uk/objects/ora:3008>
Griffin, M. J. (2012a). What Has Aristotelian Dialectic to Offer a Neopla-
tonist? A Possible Sample of Iamblichus at Simplicius on the Categories
12, 10–13, 12. International Journal of the Platonic Tradition, 6, 173–85.
Griffin, M. J. (2012b). What Does Aristotle Categorize? Semantics and the
Early Peripatetic Reading of the Categories. Bulletin of the Institute of
Classical Studies, 55, 69–108.
Griffin, M. J. (2013a). Universals, Education, and Philosophical Methodology
in Later Neoplatonism. In R. Chiaradonna and G. Galluzzo (eds), Univer-
sals in Ancient Philosophy, pp. 353–80. Pisa: Edizioni della Normale.
Griffin, M. J. (2013b). Which ‘Athenodorus’ Commented on Aristotle’s
Categories? Classical Quarterly, 63, 199–208.
Grilli, A. (1971). Un frammento d’ Aristone Alessandrino in Profirio. Gior-
nale Italiano di Filologia, 23, 292–307.
Guilhou, N. (1998). Les deux morts d’Osiris, d’après les textes des Pyramides.
Egypte, 10.
Hadot, I. (1978). Le problème du néoplatonisme alexandrin: Hiéroclès et
Simplicius. Paris: Études Augustiniennes.
 Bibliography 257
Hadot, I. (ed.). (1987). Simplicius: sa vie, son oeuvre, sa survie. Berlin: Walter
De Gruyter.
Hadot, I. (ed.). (1990a). Simplicius: Commentaire sur les Catégories. Fascicule
I—Introduction, Première Partie. (P. Hoffmann, trans.). Leiden: Brill.
Hadot, I. (1990b). Commentaire sur les Catégories (Vol. 3). Leiden: Brill.
Hadot, I. (1991). The Role of the Commentaries on Aristotle in the Teaching
of Philosophy according to the Prefaces of the Neoplatonic Commentaries
to the Categories. In H. J. Blumenthal and J. Annas (eds), Aristotle and the
Later Tradition (Oxford Studies in Ancient Philosophy, Vol. Suppl,
pp. 175–89). Oxford: Clarendon Press.
Hadot, I. (1996). Commentaire sur le manuel d’Épictète: Introduction et
edition critique du texte grec. Leiden: Brill.
Hadot, I. (2004). Simplicius, in Cat., p. 1,3–3,17 Kalbfleisch: An Important
Contribution to the History of the Ancient Commentary. Rheinisches
Museum für Philologie, 147, 408–20.
Hadot, I. (2007). Versuch einer doktrinalen Neueinordnung der Schule der
Sextier. Rheinisches Museum für Philologie, 150, 179–210.
Hadot, P. (1968). Porphyre et Victorinus. Paris: Études Augustiniennes.
Hadot, P. (1990). The Harmony of Plotinus and Aristotle According to
Porphyry. In R Sorabji (ed.), Aristotle Transformed, pp. 125–40. London/
Ithaca, NY: Duckworth/Cornell University Press.
Hadot, P. (1995). Philosophy as a Way of Life: Spiritual Exercises from
Socrates to Foucault. Oxford: Blackwell.
Hamlyn, D. W. (1977). The Presidential Address: Focal Meaning. Proceed-
ings of the Aristotelian Society, 78, 1–18.
Hamlyn, D. W. (1990). Aristotle on Dialectic. Philosophy, 65 (254), 465–76.
Harari, O. (2004). Knowledge and Demonstration: Aristotle’s Posterior Ana-
lytics. Dordrecht: Kluwer Academic Publishers.
Harari, O. (2006). Methexis and Geometrical Reasoning in Proclus’ com-
mentary on Euclid’s Elements, Oxford Studies in Ancient Philosophy, 30,
361–89.
Hatzimichali, M. (2013). The texts of Plato and Aristotle in the first century
bc. In M. Schofield (ed.), Aristotle, Plato, and Pythagoreanism in the First
Century bce, pp. 1–27. Cambridge: Cambridge University Press.
Hein, C. (1985). Definition und Einteilung der Philosophie: Von der spätan-
tiken Einleitungsliteratur zur arabischen Enzyklopädie (Europäische
Hochschulschriften. Reihe 20, Philosophie). Frankfurt: P. Lang.
Helmig, C. (2012). Forms and Concepts: Concept Formation in the Platonic
Tradition. Berlin: Walter de Gruyter.
Helmig, C., and Bonazzi, M. (eds). (2007). Platonic Stoicism—Stoic Platon-
ism: The Dialogue between Platonism and Stoicism in Antiquity. Leuven:
Leuven University Press.
258 Bibliography
Helmreich, G., Marquardt, J., and von Müller, I. (1891). Claudii Galeni
Pergameni scripta minora (Vol. 2). Leipzig: Teubner.
Henry, P. (1987). Apories orales de Plotin sur les ‘Catégories’ d’Aristote. In
J. Wiesner (ed.), Aristoteles. Werk Und Wirkung. Paul Moraux Zum 65
Geburtstag Gewidmet—Band 2: Kommentierung, Uberlieferung, Nachle-
ben, 120–56. Berlin: Walter de Gruyter.
Hijmans, B. (1975). Athenodorus on the Categories and a Pun on Atheno-
dorus. In J. Mansfeld and L. M. De Rijk (eds), Kephalaion: Studies in Greek
Philosophy and its Continuation Offered to Professor C. J. de Vogel,
pp. 105–14. Assen: Van Gorcum.
Hintikka, J. (1959). Aristotle and the Ambiguity of Ambiguity. Inquiry: An
Interdisciplinary Journal of Philosophy, 2, 137–51.
Hirsch, E. (1996). Précis of Dividing Reality. Philosophy and Phenomeno-
logical Research, 56, 199–202.
Hirsch, E. (1997). Dividing Reality. Oxford: Oxford University Press.
Hoffmann, P. (1987). Catégories et langage selon Simplicius—la question du
‘skopos’ du traité aristotélicien des Catégories. In I. Hadot (ed.), Simpli-
cius: sa vie, son oeuvre, sa survie, pp. 61–90. Berlin: Walter de Gruyter.
Hoffmann, P. (1998). La fonction des prologues exégètiques dans la pensée
pédagogique néoplatonicienne. In J.-D. Dubois and B. Roussel (eds),
Entrer en matière: Les prologues, pp. 209–45. Paris: Cerf.
Hoffmann, P. (2008). What was Commentary in Late Antiquity? The
Example of the Neoplatonic Commentators. In M. L. Gill and
P. Pellegrin (eds), A Companion to Ancient Philosophy, pp. 597–622.
New York: John Wiley & Sons.
Hubert, C. (1954). Plutarchi moralia 6.1. Leipzig: Teubner.
Huby, P. (1975). Pseudo-Archytos uber ̈ die Kategorien-Texte zur grie-
chischen Aristoteles exegese by Thomas Alexander Szlezak. Classical
Review, 25, 24–6.
Huby, P. (1981). An Excerpt from Boethus of Sidon’s Commentary on the
Categories? Classical Quarterly, 31, 398–409.
Huby, P., and Neal, G. (eds). (1989). The Criterion of Truth: Essays Written
in Honour of George Kerferd, Together With a Text and Translation (with
Annotations) of Ptolemy’s ‘On the Kriterion and Hegemonikon’. Liverpool:
Liverpool University Press.
Hülser, K. (1987). Die Fragmente zur Dialektik der Stoiker: Neue Sammlung
der Texte mit deutscher Übersetzung und Kommentaren. Stuttgart:
Frommann-Holzboog.
Husik, I. (1952). Philosophical Essays, Ancient, Mediaeval and Modern.
(M. C. Nahm and L. Strauss, eds). Oxford: Basil Blackwell.
Hyde, D. (2008). Vagueness, Logic and Ontology. Surrey, UK: Ashgate
Publishing.
 Bibliography 259
Ierodiakonou, K. (1993). The Stoic division of philosophy. Phronesis:
A Journal for Ancient Philosophy, 38, 57–74.
Ierodiakonou, K. (ed.). (1999). Topics in Stoic Philosophy. Oxford: Clarendon
Press.
Inwagen, P. van. (1995). Material Beings. Ithaca, NY: Cornell University
Press.
Inwood, B. (1985). Ethics and Human Action in Early Stoicism. Oxford:
Clarendon Press.
Inwood, B. (1993). Seneca and Psychological Dualism. In J. Brunschwig and
M. Nussbaum (eds), Passions and Perceptions, pp. 150–83. Cambridge:
Cambridge University Press.
Irwin, T. (1981). Homonymy in Aristotle. Review of Metaphysics, 34, 523–44.
Irwin, T. (1982). Aristotle’s Concept of Signification. In M. Schofield and
M. C. Nussbaum (eds), Language and Logos, pp. 241–66. Cambridge:
Cambridge University Press.
Irwin, T. (1988). Aristotle’s First Principles. Oxford: Oxford University Press.
Johnson, M. R. (2005). Aristotle on Teleology. Oxford: Oxford University
Press.
Jones, B. (1972). Individuals in Aristotle’s Categories. Phronesis, 17, 107–23.
Jones, B. (1975). An Introduction to the First Five Chapters of Aristotle’s
Categories. Phronesis, 20, 146–72.
Gentzler, J. (ed.) (1998). Method in ancient philosophy. Oxford: Clarendon
Press.
Kahn, C. (1966). The Greek Verb ‘To Be’ and the Concept of Being. Foun-
dations of Language, 2, 245–65.
Kahn, C. (1973). The Verb ‘Be’ in Ancient Greek. Indianapolis: Hackett
Publishing Co.
Kahn, C. (1981). Some Philosophical Uses of ‘To Be’ in Plato. Phronesis, 26,
105–34.
Kalbfleisch, K. (1907). Simplicii in Aristotelis categorias commentarium (CAG
8). Berlin: Reimer.
Karamanolis, G. E. (2004). Porphyry, the first Platonist Commentator of
Aristotle. In P. Adamson, H. Baltussen, and M. Stone (eds), Science and
Exegesis in Greek, Arabic and Latin, vol. 1, Supplement to the Bulletin of the
Institute of Classical Studies, vol. 83.1–2, pp. 79–113. London: Institute of
Classical Studies, University of London.
Karamanolis, G. E. (2006). Plato And Aristotle in Agreement?: Platonists on
Aristotle from Antiochus to Porphyry. Oxford: Clarendon Press.
Karamanolis, G. E. (2010). Plutarch. The Stanford Encyclopedia of Philo-
sophy (Fall 2010 Edition) E. Zalta (ed.). <http://plato.stanford.edu/entries/
plutarch/>
Karamanolis, G., and Sheppard, A. (eds). (2007). Studies on Porphyry.
London: Institute of Classical Studies, University of London.
260 Bibliography
Keefe, R., and Smith, P. (eds). (1999). Vagueness: A Reader. Cambridge, MA:
MIT Press.
Kerferd, G. B. (1981). The Sophistic Movement. Cambridge: Cambridge
University Press.
King, D. (2010). The Earliest Syriac Translation of Aristotle’s Categories: Text,
Translation and Commentary. Leiden: Brill.
Kneale, W., and Kneale, M. (1984). The Development of Logic. Oxford:
Clarendon Press.
Konstan, D. (2007). Aspasius: On Aristotle Nicomachean Ethics 1–4, 7–8.
London: Duckworth.
Krafft, P. (1975). Die handschriftliche Überlieferung von Cornutus’ Theologia
Graeca. Heidelberg: Winter.
Kraut, R. (2013). Plato. The Stanford Encyclopedia of Philosophy (Fall 2013
Edition) E. Zalta (ed.). <http://plato.stanford.edu/entries/plato/>
Kretzmann, N. (1974). Aristotle on Spoken Sound Significant by Conven-
tion. In J. Corcoran (ed.), Ancient Logic and Its Modern Interpretations,
Vol. 9, pp. 3–21. Dordrecht: D. Reidel.
Larsen, B. D. (1972). Jamblique de Chalcis. Exégète et Philosophe. Aarhus:
Universitetsforlaget.
Lewis, F. A. (1991). Substance and Predication in Aristotle. Cambridge:
Cambridge University Press.
Littig, F. (1890). Andronicus von Rhodos I. Munich: F. Straub.
Littig, F. (1894). Andronicus von Rhodos II. Erlangen.
Littig, F. (1895). Andronicus von Rhodos III. Erlangen.
Lloyd, A. C. (1952). Plato’s description of division. The Classical Quarterly, 2,
105–12.
Lloyd, A. C. (1955). Neoplatonic Logic and Aristotelian Logic—I. Phronesis,
1, 58–72.
Lloyd, A. C. (1962). Genus, species and ordered series in Aristotle. Phronesis,
7, 67–90.
Lloyd, A. C. (1978). Emotion and Decision in Stoic Psychology. In J. Rist
(ed.), The Stoics, pp. 233–46. Berkeley: University of California Press.
Lloyd, A. C. (1990). The Anatomy of Neoplatonism. Oxford: Clarendon Press.
Lloyd, G. E. R. (1990). Plato and Archytas in Plato’s Seventh Letter. Phron-
esis, 35, 159–74.
Long, A. A. (ed.). (1971). Problems in Stoicism. London: Athlone Press.
Long, A. A. (1978). Dialectic and the Stoic Sage. In J. Rist (ed.), The Stoics,
pp. 101–24. Berkeley: University of California Press.
Long, A. A. (1989). Ptolemy on the Criterion: An Epistemology for the
Practising Scientist. In P. Huby and G. Neal (eds), The Criterion of
Truth, pp. 151–78. Liverpool: Liverpool University Press.
Long, A. A. (1996). Stoic Studies. Cambridge: Cambridge University Press.
 Bibliography 261
Long, A. A. (2005). Stoic Linguistics, Plato’s ‘Cratylus,’ and Augustine’s ‘De
Dialectica’. In D. Frede and B. Inwood (eds), Language and Learning:
Philosophy of Language in the Hellenistic Age, pp. 36–55. Cambridge:
Cambridge University Press.
Long, A. A., and Sedley, D. (eds). (1987). The Hellenistic Philosophers.
Cambridge: Cambridge University Press.
Long, H. S. (1964). Diogenis Laertii vitae philosophorum. Oxford: Clarendon
Press.
Louis, P. (ed.) (1945). Albinos. Épitomé. Paris: Les Belles Lettres.
Loux, M. J. (1991). Primary Ousia: An Essay on Aristotle’s Metaphysics Z and
H. Ithaca, NY: Cornell University Press.
Loux, M. J. (2006). Metaphysics: A Contemporary Introduction. (3rd ed.).
Abingdon, UK: Routledge.
Lowe, E. J. (2006). The Four-Category Ontology: A Metaphysical Foundation
for Natural Science. New York: Oxford University Press, USA.
Łukasiewicz, J. (1935). Zur Geschichte der Aussagenlogik. Erkenntnis, 5,
111–31.
Łukasiewicz, J. (1951). Aristotle’s Syllogistic from the Standpoint of Modern
Formal Logic. Oxford: Clarendon Press.
Luna, C. (2001a). Trois études sur la tradition des commentaires anciens à la
Métaphysique d’Aristotle. Leiden: Brill.
Luna, C. (2001b). Commentaire. In P. Hoffmann and C. Luna (eds), Simpli-
cius: Commentaire sur les Categories d’Aristote. Chapitres 2–4. Paris: Les
Belles Lettres.
Luna, C. (2013). Boéthos de Sidon sur les relatifs. Studia graeco-arabica 3,
1–35.
Lynch, J. P. (1972). Aristotle’s School: A Study of Greek Educational Institu-
tion. Berkeley: University of California Press.
Maas, M. (ed.). (2005). The Cambridge Companion to the Age of Justinian.
Cambridge: Cambridge University Press.
Mackay, E. A. (2008). Orality, literacy, memory in the ancient Greek and
Roman world. Leiden: Brill.
Madigan, A. (1994). Alexander on Species and Genera as Principles. In
L.P. Schrenk (ed.), Aristotle in Late Antiquity, pp. 74–91. Washington,
DC: Catholic University of America Press.
Magee, J. (1989). Boethius on Signification and Mind (Vol. 52). Leiden: Brill.
Magee, J. (1998). Anicii Manlii Severini Boethii De Divisione Liber. Leiden:
Brill.
Mahé, J. P. (1990). David l’Invincible dans la tradition arménienne. In
I. Hadot (ed.), Simplicius Commentaire sur les Catégories Vol. 50,
pp. 189–207. Leiden: Brill.
Maier, H. (1896). Die Syllogistik des Aristoteles (Vol. 1–3). Tübingen:
H. Laupp’schen.
262 Bibliography
Mann, W.-R. (2000). The Discovery of Things: Aristotle’s Categories & Their
Context. Princeton: Princeton University Press.
Mansfeld, J. (1992). Heresiography in Context: Hippolytus’ Elenchos as a
Source for Greek Philosophy. Leiden: Brill.
Mansfeld, J. (1994). Prolegomena: Questions to Be Settled before the Study of
an Author, or a Text (Vol. 61). Leiden: Brill.
Mansfeld, J., and De Rijk, L. M. (eds). (1975). Kephalaion: Studies in Greek
Philosophy and its Continuation Offered to Professor C. J. de Vogel. Assen:
Van Gorcum.
Mansfeld, J., and Runia, D. T. (1997). Aëtiana: The Method and Intellectual
Context of a Doxographer—Volume I: The Sources (Vol. 73). Leiden: Brill.
Mansion, S. (1949). La doctrine aristotélicienne de la substance et le Traité des
Catégories. Proceedings of the Tenth International Congress of Philoso-
phy. Amsterdam.
Mariette, A. (1873). Dendérah (Vol. 4). Paris: Librairie A. Franck.
Mariotti, I. (1966). Aristone d’Alessandria. Bologna: Pàtron Editore.
Marquardt, J., von Müller, I., and Helmreich, G. (1884). Claudii Galeni
Pergameni Scripta Minora (Vol. 1). Leipzig: Teubner.
Mason, H. J. (1983). The Distinction of Lucius in Apuleius’ ‘Metamorph-
oses’. Phoenix, 37, 135–43.
Matthen, M. (1983). Greek Ontology and the ‘Is’ of Truth. Phronesis, 28,
113–35.
Mau, J., and Mutschmann, H. (1914). Sexti Empirici opera, vol. 2. Leipzig:
Teubner.
Mazzarelli, C. (1985). Raccolta e interpretazione delle testimonianze e dei
frammenti del medioplatonico Eudoro di Allessandria (in due parti).
Rivista di fiolosofia neo-scolastica, 77, 197–209; 535–55.
McGinnis, J., and Reisman, D. (trs). (2007). Classical Arabic Philosophy: An
Anthology of Sources. Indianapolis: Hackett Publishing Co.
Mcguire, J. E., and Strange, S. K. (1988). An Annotated Translation of
Plotinus Ennead iii 7: On Eternity and Time. Ancient Philosophy, 7,
251–71.
McKirahan, R. (2001). Aristotle’s Theory of Predication. Apeiron: A Journal
for Ancient Philosophy and Science, 34, 321–8.
McPherran, M. L. (2007). Socratic Epagōgē and Socratic Induction. Journal
of the History of Philosophy, 45, 347–64.
Meeks, D. (2006). Mythes et légendes du Delta: D’après le papyrus Brooklyn
47.218.84. Le Caire: Institut français d’archéologie orientale du Caire.
Meeks, D., and Favard-Meeks, C. (1996). Daily Life of the Egyptian Gods.
(G.M. Goschgarian trans.). Ithaca, NY: Cornell University Press.
Meineke, A. (1859). Zu Stobaeus. Zeitschrift für das Gymnasialwesen, 13,
563–65.
 Bibliography 263
Mendelson, E. (1997). Introduction to Mathematical Logic (4th ed.). London:
Chapman & Hall.
Menn, S. (1995). Metaphysics, Dialectic and the Categories. Revue de Méta-
physique et de Morale 3, 311–37.
Menn, S. (1999). The Stoic Theory of Categories. Oxford Studies in Ancient
Philosophy, 17, 215–48.
Menn, S. (forthcoming). A Note on the Title of Porphyry’s Categories
Commentary —æ‘  ˆÆºØ.
Merlan, P. (1960). From Platonism to Neoplatonism. The Hague: Martinus
Nijhoff.
Merlan, P. (1967). The Peripatos. In A. H. Armstrong (ed.), The Cambridge
History of Later Greek and Early Medieval Philosophy, pp. 107–23.
Cambridge: Cambridge University Press.
Migne, J.-P. (1891). Anicii Manlii Severini Boethii In Categorias Aristotelis.
Paris: Garnier.
Moraux, P. (1942). Alexandre d'Aphrodise: Exégète de la noétique d'Aristote.
Liège/Paris: Faculté de philosophie et lettres/E. Droz.
Moraux, P. (1951). Les listes anciennes des ouvrages d’Aristote. Louvain:
Éditions universitaires.
Moraux, P. (1967). Aristoteles, der Lehrer Alexanders von Aphrodisias.
Archiv für Geschichte der Philosophie, 42, 169–82.
Moraux, P. (1969) ‘Eine Korrektur des Mittelplatonikers Eudoros zum Text
der Metaphysik des Aristoteles’. In R. Stiehl and H.E. Stier (eds). Beiträge
zur alten Geschichte und deren Nachleben, pp. 492–504. Berlin: Walter de
Gruyter.
Moraux, P. (1973). Der Aristotelismus bei den Griechen von Andronikos bis
Alexander von Aphrodisias—Erster Band: Die Renaissance des Aristotelis-
mus im I. Jh. v. Chr. (Vol. 1/= Peripatoi 5). Berlin/New York: Walter de
Gruyter.
Moraux, P. (1984). Der Aristotelismus bei den Griechen von Andronikos bis
Alexander von Aphrodisias—Zweiter Band: Der Aristotelismus im I. und
II. Jh. n.Chr (Vol. 2/= Peripatoi 6). Berlin/New York: Walter de Gruyter.
Morison, B. (2005). Les Catégories d’Aristote comme introduction à la
logique. In O. Bruun and L. Corti (eds), Les Catégories et leur histoire,
pp. 103–20. Paris: Vrin.
Most, G. W. (1989). Cornutus and Stoic Allegoresis: A Preliminary Report.
In W. Haase (ed.), Aufstieg und Niedergang der römischen Welt, Vol. 36.3,
pp. 2014–65. Berlin: Walter de Gruyter.
Most, G. W. (ed.). (1999). Commentaries—Kommentare. (Vol. 4). Göttingen:
Vandenhoeck & Ruprecht.
Mueller, I. (1978). An Introduction to Stoic Logic. In J. Rist (ed.), The Stoics,
pp. 1–26. Berkeley: University of California Press.
264 Bibliography
Mullach, F. W. A. (1867). Fragmenta Philosophorum Graecorum. Paris:
Firmin Didot.
Napolitano, L.M. (1985). ‘Eudoro di Alessandria: Monismo, dualismo, assi-
ologia dei principi nella tradizione Platonica’, Museum Patavinum 3,
289–312.
Natali, C. and Hutchinson. D.S. (2013). Aristotle: His Life and School.
Princeton: Princeton University Press.
Newton, L. A. (ed.). (2008). Medieval Commentaries on Aristotle’s Categories.
Leiden: Brill.
Nock, A.D. (1931). ‘Kornutos’. RE Suppl. 5 (pp. 995–1005). Stuttgart.
Nussbaum, M. C. (1982). Saving Aristotle’s Appearances. In M. Schofield
and M. C. Nussbaum (eds), Language and Logos, pp. 267–95. Cambridge:
Cambridge University Press.
Nutton, V. (ed.). (1981). Galen: Problems and Prospects. London: Wellcome
Institute for the History of Medicine.
Obbink, D. (1992). ‘What all men believe—Must be true’: Common Con-
ceptions and consensio omnium in Aristotle and Hellenistic Philosophy.
Oxford Studies in Ancient Philosophy, 10, 193–231.
Oderberg, D. S. (2005). The Old New Logic: Essays On The Philosophy Of Fred
Sommers. Cambridge, MA: MIT Press.
Oehler, K. (1984). Aristoteles Kategorien; übersetzt und erläutert von Klaus
Oehler (Vol. 1.1). Berlin: Akadamie-Verlag.
Oliver, J. H. (1981). Marcus Aurelius and the Philosophical Schools at
Athens. The American Journal of Philology, 102, 213–25.
Opsomer, J., and Sharples, B. (2000). Alexander of Aphrodisias, ‘De Intel-
lectu’ 110.4: ‘I Heard This from Aristotle’. A Modest Proposal. The
Classical Quarterly, 50, 252–6.
Owen, G. E. L. (1960). Logic and Metaphysics in Some Earlier Works of
Aristotle. In I. Düring and G. E. L. Owen (eds), Aristotle and Plato in the
Mid-Fourth Century, pp. 163–90. Göteborg: Almqvist and Wiksell.
Owen, G. E. L. (1965). Inherence. Phronesis, 10, 97–105.
Patzig, G. (1968). Aristotle’s Theory of Syllogism: A Logico-Philological Study
of Book A of the Prior Analytics. (J. Barnes, trans.). Dordrecht: D. Reidel.
Pépin, J. (1980). Clément d’Alexandrie, les Catégories d’Aristote et le frag-
ment 60 d’Héraclite. In P. Aubenque (ed.), Concepts et catégories dans la
pensée antique, pp. 271–84. Paris: Vrin.
Perin, C. (2007). Substantial Universals in Aristotle’s Categories. Oxford
Studies in Ancient Philosophy, 33, 125–44.
Perler, D. (ed.). (2001). Ancient and Medieval Theories of Intentionality.
Leiden: Brill.
Phillips, J. (1987). Stoic ‘Common Notions’ in Plotinus. Dionysius, 11, 33–52.
Pinborg, J. (1962). Das Sprachdenken der Stoa und Augustins Dialektik.
Classica et Mediaevalia, 23, 148–77.
 Bibliography 265
Pines, S. (1961). Omne quod movetur necesse est ab aliquo moveri:
A Refutation of Galen by Alexander of Aphrodisias and the Theory of
Motion. Isis, 52, 21–54.
Plezia, M. (1946). De Andronici Rhodii studiis aristotelicis. Kraków: Polskiej
Akademii Umiejȩtności, Skkad ggówny w ksiȩg, Gebethnera i Wolffa.
Plezia, M. (1961). Epistularum fragmenta cum Testamento. Warsaw: Państ-
wowe Wydawn Naukowe.
Praechter, K. (1922). Nikostratos der Platoniker. Hermes, 57.4, 481–517.
Praechter, K. (1973). Kleine Schriften, edited by H. Dörrie. Hildesheim:
Georg Olms Verlag.
Prantl, C. (1846). Zeitschrift für die Alterhthumswissenschaft 4, 641–52.
Prantl, C. (1855). Geschichte der Logik im Abenlande. Vol. 1. Leipzig: S. Hirzl.
Repr. Hildesheim: Georg Olms, 1997.
Primavesi, O. (2007). Ein Blick in den Stollen von Skepsis: Vier Kapitel zur
frühen Überlieferung des Corpus Aristotelicum, Philologus, 151, 51–77.
Quine, W. V. (1960). Word and Object. Cambridge, MA: MIT Press.
Quine, W. V. (1981). What Price Bivalence? Journal of Philosophy, 78, 90–5.
Radt, S. L. (1988). ˇd (Æd etc.) æd + acc. nominis proprii bei Strabon.
Zeitschrift für Papyrologie und Epigraphik, 71, 35–40.
Rappe, S. (2000). Reading Neoplatonism: Non-discursive Thinking in
the Texts of Plotinus, Proclus, and Damascius. Cambridge: Cambridge
University Press.
Rashed, M. (2004). Priorité de l’r  ou du ª  entre Andronicos et
Alexandre: vestiges arabes et grecs inédits. Arabic Sciences and Philosophy,
14, 9–63.
Rashed, M. (2005). Aristote. De la géneration et la corruption. Paris: Les
Belles Lettres.
Rashed, M. (2007). Essentialisme: Alexandre D’Aphrodise Entre Logique,
Physique Et Cosmologie. Berlin: Walter de Gruyter.
Rashed, M. (2013). Boethus’ Aristotelian Ontology. In M. Schofield (ed.),
Aristotle, Plato, and Pythagoreanism in the First Century bce, pp. 53–77.
Cambridge: Cambridge University Press.
Rehm, A. (1927). Sosigenes [6]. Paulys Realencyclopädie der classischen
Altertumswissenschaft. III.A.1. 1153–7.
Rehn, R. (2000). Sprache und Dialektik in der Aristotelischen Philosophie
(Vol. 31). Amsterdam/Philadelphia, PA: B.R. Grüner.
Reinhardt, T. (2003). Cicero’s Topica. Critical edition, translation, introduc-
tion, and commentary. Oxford: Oxford University Press.
Reinhardt, T. (2007). Andronicus of Rhodes and Boethus of Sidon on
Aristotle’s ‘Categories’. In R. W. Sharples and R. Sorabji (eds), Greek
266 Bibliography
and Roman Philosophy, 100 bc–200 ad, Vol. 2, pp. 513–29. London:
Institute of Classical Studies, University of London.
Reinhardt, T. (2008). Epicurus and Lucretius on the Origins of Language.
The Classical Quarterly, 58, 127–40.
Reisman, D. C., and Taylor, R. C. (2005). Al-Fārābī and the Philosophical
Curriculum. In P. Adamson (ed.), (pp. 52–71). Cambridge: Cambridge
University Press.
Renon, L.V. (1998). Aristotle’s Endoxa and Plausible Argumentation. Argu-
mentation, 12, 95–113.
Rescher, R., and Marmura, M. E. (1965). The Refutation of Alexander of
Aphrodisias of Galen’s Treatise of the Theory of Motion. Islamabad: Islamic
Research Institute.
Rist, J. (1971). Categories and Their Uses. In A. A. Long (ed.), Problems in
Stoicism, pp. 38–57. London: Athlone Press.
Rist, J. (ed.). (1978). The Stoics. Berkeley: University of California Press.
Rist, J. (1986). ‘Scepticism or Platonism? The Philosophy of the Fourth
Academy’ by Harold Tarrant. Review by John M. Rist. Phoenix, 40, 467–9.
Rist, J. (1989a). The Mind of Aristotle: A Study in Philosophic Growth.
Toronto: University of Toronto Press.
Rist, J. (1989b). Seneca and Stoic Orthodoxy. In W. Haase (ed.), Aufstieg
und Niedergang der römischen Welt, Vol. 36.3, pp. 1993–2012. Berlin/
New York: Walter de Gruyter.
Rohde, E. (1925). Psyche: The Cult of Souls and the Belief in Immortality
Among the Greeks (Vol. 6). London: Routledge.
Rose, V. (1854). De Aristotelis librorum ordine et auctoritate. Berlin: Reimer.
Ross, W. D. (1924). Aristotle’s Metaphysics. Oxford: Clarendon Press.
Rubarth, S. (2011). Review of: Prolepsis and Ennoia in the Early Stoa by
H. Dyson, Bryn Mawr Classical Review 2011.01.38
Russell, B. (1905). On Denoting. Mind, 14, 479–93.
Russell, B. (1912). The Problems of Philosophy. London: Williams & Norgate.
Russell, B., and Whitehead, A. N. (1910). Principia Mathematica. Cam-
bridge: Cambridge University Press.
Ryle, G. (1939). Plato’s Parmenides (II.). Mind, 48, 302–25.
Ryle, G. (1961). Categories. In A. Flew (ed.), Logic and Language, pp. 65–81.
Oxford: Basil Blackwell.
Ryle, G. (1966). Plato’s Progress. Cambridge: Cambridge University Press.
Sachs, D. (1948). Does Aristotle Have a Doctrine of Secondary Substances?
Mind, 57, 221–5.
Salles, R. (ed.). (2005). Metaphysics, Soul and Ethics in Ancient Thought:
Themes from the Work of Richard Sorabji. Oxford: Oxford University
Press.
Sandbach, F. H. (1971). Ennoia and Prolepsis in the Stoic Theory of Know-
ledge. Classical Quarterly, 24, 44–51.
 Bibliography 267
Sandbach, F. H. (1985). Aristotle and the Stoics (Vol. Suppl). Cambridge:
Cambridge Philological Society.
Schmidt, E. (1966). Alexander von Aphrodisias in einem altarmenischen
Kategorien-Kommentar. Philologus, 110, 277–86.
Schmidt, H. (1907). De Hermino Peripatetico. Marburg: Bauer.
Schofield, M. (ed.). (2013). Aristotle, Plato and Pythagoreanism in the First
Century bc: New Directions for Philosophy. Cambridge: Cambridge Uni-
versity Press.
Schofield, M., and Nussbaum, M. C. (1982). Language and Logos: Studies in
Ancient Greek Philosophy Presented to G. E. L. Owen. Cambridge: Cam-
bridge University Press.
Schrenk, L.P. (1994). Aristotle in Late Antiquity. Washington, DC: Catholic
University of America Press.
Sedley, D. (1982). The Stoic Criterion of Identity. Phronesis, 27, 255–75.
Sedley, D. (1997). Plato’s Auctoritas and the Rebirth of the Commentary
Tradition. In J. Barnes and M. Griffin (eds), Philosophia Togata II: Plato
and Aristotle at Rome, pp. 110–29. Oxford: Oxford University Press.
Sedley, D. (1998). The Etymologies in Plato’s Cratylus. The Journal of
Hellenic Studies, 118, 140–54.
Sedley, D. (2002). Aristotelian Relativities. In M. Canto-Sperber and
P. Pellegrin (eds), Le style de la pensée: Recueil de textes en hommage à
Jacques Brunschwig, pp. 324–52. Paris: Les Belles Lettres.
Sedley, D. (2005). Stoic Metaphysics at Rome. In Ricardo Salles (ed.), Meta-
physics, Soul and Ethics in Ancient Thought: Themes from the Work of
Richard Sorabji, pp. 117–42. Oxford: Oxford University Press.
Sedley, D. (ed.). (2012). The Philosophy of Antiochus. Cambridge: Cambridge
University Press.
Sellars, J. (2004). The Aristotelian Commentators: A Bibliographical Guide.
In P. Adamson, H. Baltussen, and M. W. F. Stone (eds), Philosophy,
Science, and Exegesis in Greek, Arabic, and Latin Commentaries, (Vol. 1,
pp. 239–68). London: Institute of Classical Studies, University of London.
Sharples, R. W. (1987). Alexander of Aphrodisias: Scholasticism and Innov-
ation. In W. Haase (ed.), Aufstieg und Niedergang der römischen Welt,
Vol. 36.2, pp. 1176–243. Berlin/New York: Walter de Gruyter.
Sharples, R. W. (1990). The School of Alexander? In R. Sorabji (ed.),
Aristotle Transformed, pp. 83–112. London/Ithaca, NY: Duckworth/
Cornell University Press.
Sharples, R. W. (1997). The Peripatetic School. In D. Furley (ed.), From
Aristotle to Augustine, Vol. 2, pp. 147–87. London: Routledge.
Sharples, R. W. (2005). Implications of the New Alexander of Aphrodisias
Inscription. Bulletin of the Institute of Classical Studies, 48, 47–56.
268 Bibliography
Sharples, R. W. (2008). Habent Sua Fata Libelli: Aristotle’s ‘Categories’ in the
First Century bc. Acta Antiqua Academiae Scientiarum Hungaricae, 48,
273–87.
Sharples, R. W. (2010). Peripatetic Philosophy, 200 bc to ad 200: An Intro-
duction and Collection of Sources in Translation. Cambridge: Cambridge
University Press.
Shiel, J. (1990). Boethius’ Commentaries on Aristotle. In R. Sorabji (ed.),
Aristotle Transformed, pp. 349–72. London/Ithaca, NY: Duckworth/Cor-
nell University Press.
Shields, C. J. (1999). Order in multiplicity: homonymy in the philosophy of
Aristotle (p. 290). Oxford: Oxford University Press.
Shields, C. J. (2003). Classical Philosophy: A Contemporary Introduction.
London: Routledge.
Sidwell, K. (2004). Lucian: Chattering Courtesans and Other Sardonic
Sketches. London: Penguin.
Sirkel, R. (Forthcoming). ‘Ontological Priority in Aristotle’s Categories’.
Sluiter, I. (1999). Commentaries and the Didactic Tradition. In G. W. Most
(ed.), Commentaries—Kommentare, Vol. 4, pp. 173–205. Göttingen: Van-
denhoeck & Ruprecht.
Smiley, T. J. (1973). What is a Syllogism? Journal of Philosophical Logic, 2,
136–54.
Smith, A. (ed.). (1993). Porphyrii philosophi fragmenta. Stuttgart: B.G. Teubner.
Smith, A. (1997). Aristotle: Topics, Books I and VIII. New York: Oxford
University Press, USA.
Smith, A. (1999). Dialectic and method in Aristotle. From Puzzles to Prin-
ciples: Essays on Aristotle’s Dialectic, 39–55. New York: Lexington Books.
Smith, A. (2010). Boethius: On Aristotle ‘On Interpretation’. London:
Duckworth.
Sorabji, R. R. K. (1983). Time, Creation, and the Continuum: Theories in
Antiquity and the Middle Ages. London: Duckworth.
Sorabji, R. R. K. (1990). Aristotle Transformed: The Ancient Commentators
and Their Influence. London: Duckworth.
Sorabji, R. R. K. (2000). Emotion and Peace of Mind: From Stoic Agitation to
Christian Temptation. Oxford: Oxford University Press.
Sorabji, R. R. K. (2005a). The Philosophy of the Commentators 200–600 ad:
A Sourcebook. Vol. 1: Psychology (With Ethics and Religion). London/
Ithaca, NY: Duckworth/Cornell University Press.
Sorabji, R. R. K. (2005b). The Philosophy of the Commentators 200–600 ad:
A Sourcebook. Vol. 2: Physics. London/Ithaca, NY: Duckworth/Cornell
University Press.
 Bibliography 269
Sorabji, R. R. K. (2005c). The Philosophy of the Commentators 200–600 ad:
A Sourcebook. Vol. 3: Logic and Metaphysics. London/Ithaca, NY: Duck-
worth/Cornell University Press.
Sorabji, R. R. K. (2007a). Modifications to Aristotle’s Physics by Peripatetics
and Others. In R. Sorabji and R. W. Sharples (eds), Greek and Roman
Philosophy 100 bc–200 ad, Vol. 2, pp. 575–94. London: Institute of
Classical Studies, University of London.
Sorabji, R. R. K. (2007b). Time, Place, and Extracosmic Place: Peripatetics in
the First Century bc and a Stoic Opponent. In R. Sorabji and
R. W. Sharples (eds), Greek and Roman Philosophy 100 bc–200 ad,
Vol. 2, pp. 563–74. London: Institute of Classical Studies, University of
London.
Sorabji, R., and Sharples, R. W. (eds). (2007). Greek and Roman Philosophy
100 bc–200 ad. London: Institute of Classical Studies, University of
London.
Spengel, L. (1845). Münchener Gelehrte Anzeigen. 20.5, c. 33–56.
Spruit, L. (1995). Species Intelligibilis: From Perception to Knowledge. Leiden:
Brill.
Stamatellos, G. (2007). Plotinus and the Presocratics: A Philosophical Study of
Presocratic Influences in Plotinus’ Enneads. Albany, NY: SUNY Press.
Strange, S. (1987). Plotinus, Porphyry, and the Neoplatonic Interpretation of
the ‘Categories.’ In W. Haase (ed.), Aufstieg und Niedergang der römischen
Welt, Vol. 36.2, pp. 955–74. Berlin/ New York: Walter de Gruyter.
Strange, S. (1992). Porphyry: On Aristotle ‘Categories’. London: Duckworth.
Strange, S. K. (1994). Plotinus on the Nature of Eternity and Time. In
L.P. Schrenk (ed.), Aristotle in Late Antiquity, pp. 22–53. Washington,
DC: Catholic University of America Press.
Striker, G. (1991). Following Nature: A Study in Stoic Ethics. Oxford Studies
in Ancient Philosophy, 9, 1–73.
Strobel, B. (2009). Von einem Subjekt ausgesagt werden und an einem
Subjekt vorliegen: zur Semantik genereller Terme in der aristotelischen
Kategorienschrift. Phronesis, 54, 40–75.
Studtmann, P. (2013). Aristotle’s Categories. The Stanford Encyclopedia
of Philosophy (Winter 2013 Edition) E. Zalta (ed.). <http://plato.stanford
.edu/entries/aristotle-categories/>
Szlezák, T. A. (ed.). (1972). Pseudo-Archytas Über die Kategorien: Texte Zur
Griechischen Aristoteles-Exegese. Berlin: Walter de Gruyter.
Tarán, L. (1978a). Anonymous Commentary on Aristotle’s ‘De Interpreta-
tione’: Codex Parisinus Graecus 2064. Messenheim am Glan: A. Hain.
Tarán, L. (1978b). Speusippus and Aristotle on Homonymy and Synonymy.
Hermes, 106, 73–99.
270 Bibliography
Tarán, L. (1981). Review of ‘Der Aristotelismus bei den Griechen von
Andronikos bis Alexander von Aphrodisias. 1. Band: Die Renaissance
des Aristotelismus im 1. Jh. v. Chr.’ by Paul Moraux. Gnomon, 53, 721–50.
Tarán, L. (2001). Collected Papers: 1962–1999. Leiden: Brill.
Tarrant, H. (1985). Scepticism or Platonism? The Philosophy of the Fourth
Academy. Cambridge: Cambridge University Press.
Tarrant, H. (1987). Peripatetic and Stoic Epistemology in Boethus and
Antiochus. Apeiron, 20, 17–37.
Tarrant, H. (2008). Eudorus and the Early Platonist Interpretation of the
‘Categories’. Laval Philosophique et Théologique, 64, 583–95.
Temporini, H., and Haase, W. (1987). Aufstieg und Niedergang der rö-
mischen Welt: (ANRW): Geschichte und Kultur Roms im Spiegel der
neueren Forschung. Principat; 36. Philosophie, Wissenschaften, Technik; 2,
Philosophie (Platonismus, Aristotelismus). Berlin: Walter de Gruyter.
Thesleff, H. (ed). (1965). The Pythagorean Texts of the Hellenistic Period =
Acta Academiae Aboensis, 30. Abo: Abo Akademi.
Thiel, R. (2004). Aristoteles’ Kategorienschrift in ihrer antiken Kommentier-
ung. Tübingen: Mohr Siebeck.
Thillet, P. (1984). Alexandre d’Aphrodise: Traité du destin. Paris: Les Belles
Lettres.
Tieleman, T. (1976). Alexander of Aphrodisias on Stoic Physics: A Study of the
‘De Mixtione’, with Preliminary Essays, Text, Translation and Commen-
tary. Leiden: Brill.
Tieleman, T. (1996). Galen and Chrysippus on the Soul: Argument and
Refutation in the ‘De Placitis’ II–III. Leiden: Brill.
Tieleman, T. (2003). Chrysippus’ ‘On Affections’: Reconstruction and Inter-
pretation. Leiden: Brill.
Todd, R. B. (1973). The stoic common notions: A re-examination and
reinterpretation. Symbolae Osloenses, 48, 47–75.
Tomasello, M., and Carpenter, M. (2007). Shared intentionality. Develop-
mental Science, 10, 121–5.
Too, Y- L. (2010). The Idea of the Library in the Ancient World. Oxford:
Oxford University Press.
Trapp, M. B. (2007). Philosophy in the Roman Empire: Ethics, Politics and
Society. Surrey, UK: Ashgate Publishing.
Tselemanis, P. (1985). Theory of Meaning and Signification in Aristotle In
K. Voudouris (ed.) Language and Reality in Greek Philosophy, pp. 194–9.
Athens: Hellenike Philosophike Hetaireia.
Tuominen, M. (2009). The Ancient Commentators on Plato and Aristotle.
Stocksfield: Acumen.
Turner, J. D. (2001). Sethian Gnosticism and the Platonic Tradition. Quebec/
Leuven: Presses de l’Université Laval/Peeters.
 Bibliography 271
Tweedale, M. M. (1984). Alexander of Aphrodisias’ views on universals.
Phronesis, 29, 279–303.
Tye, M. (1990). Vague Objects. Mind, 99, 535–57.
Unger, P. (1979). There Are No Ordinary Things. Synthese, 41, 117–54.
Van den Berg, R. M. (2008). Proclus’ ‘Commentary on the Cratylus’ in
Context: Ancient Theories of Language and Naming. Leiden: Brill.
Van den Berg, R. M. (2009). As we are always speaking of them and using
their names on every occasion. Plotinus, Enn. III.7 [45]: language, experi-
ence and the philosophy of time in Neoplatonism. In R. Chiaradonna and
F. Trabbatoni (eds), Physics and Philosophy of Nature in Greek Neopla-
tonism: Proceedings of the European Science Foundation Exploratory
Workshop, pp. 101–20. Leiden: Brill.
Vander Waerdt, P. A. (ed.). (1994). The Socratic Movement. Ithaca, NY:
Cornell University Press.
Verlinsky, A. (2005). Epicurus and His Predecessors on the Origin of Lan-
guage. In D. Frede and B. Inwood (eds), Language and Learning: Philoso-
phy of Language in the Hellenistic Age, pp. 56–100. Cambridge: Cambridge
University Press.
Vlastos, G. (1973). An Ambiguity in the ‘Sophist’. In G. Vlastos (ed.) Platonic
Studies, pp. 270–322. Princeton: Princeton University Press.
Vlastos, G. (1974). A Note on ‘Pauline Predictions’ in Plato. Phronesis, 19,
95–101.
Von Fritz, K. (1936). Nikostratos 26. RE 17.1. Cols 547–51.
Von Müller, I. (1897). Über Galens Werk vom wissenschaftlichen Beweis.
In Abhandlungen der Königlich Bayerischen Akademie der Wissenschaften
Philosophisch-Historische Klasse, Vol. 20, pp. 403–78. Munich: K. Akademie,
in Kommission des G. Franz’schen Verlags.
Watts, E. (2008). City and School in Late Antique Athens and Alexandria.
Berkeley: University of California Press.
Wear, S. K. (2008). Syrianus the Platonist on Eternity and Time. Classical
Quarterly, 58, 648–60.
Wedin, M. (2000). Aristotle’s Theory of Substance: The Categories and Meta-
physics Zeta. Oxford: Oxford University Press.
Wehrli, F. (1967). Die Schule des Aristoteles (2nd ed.). Basel: Schwabe.
Weidemann, H. (1994). Aristoteles ‘Peri Hermeneias’; übersetzt und erläutert
von Hermann Weidemann, Vol. 1.2. Berlin: Akadamie-Verlag.
Westerink, L. G. (1961). Elias on the Prior Analytics. Mnemosyne, 14,
126–39.
Westerink, L. G. (1990). The Alexandrian Commentators and the Introduc-
tions to Their Commentators. In R. Sorabji (ed.), Aristotle Transformed,
pp. 325–48. London/Ithaca, NY: Duckworth/Cornell University Press.
Whitaker, C. W. A. (1996). Aristotle’s ‘De Interpretatione’: Contradiction and
Dialectic. Oxford: Oxford University Press.
272 Bibliography
Whitman, J. (1987). Allegory: The Dynamics of an Ancient and Medieval
Technique. Cambridge, MA: Harvard University Press.
Whittaker, J. (ed) (1990). Alcinoos: Enseignement des doctrines de Platon.
Introduction, texte établi et commenté par J. Whittaker et traduit par
P. Louis. Paris: Belles Lettres.
Wiesner, J. (ed.). (1987). Aristoteles Werk und Wirkung, Paul Moraux
gewidmet: Zweiter Band: Kommentierung, Überlieferung, Nachleben. Ber-
lin: Walter de Gruyter.
Wilberding, J. (2005). Aristotle, Plotinus, and Simplicius on the Relation of
the Changer to the Changed. Classical Quarterly, 55, 447–54.
Wildberg, C. (1990). Three Neoplatonic Introductions to Philosophy: Am-
monius, David, and Elias. Hermathena, 140, 33–51.
Wildberg, C. (2005). Philosophy in the Age of Justinian. In M. Maas (ed.),
The Age of Justinian, pp. 316–40. Cambridge: Cambridge University Press.
Wildberg, C. (2008a). David. The Stanford Encyclopedia of Philosophy (Fall
2008 Edition) E. Zalta (ed.). <http://plato.stanford.edu/entries/david>
Wildberg, C. (2008b). Olympiodorus. The Stanford Encyclopedia of Philoso-
phy (Fall 2008 Edition) E. Zalta (ed.). <http://plato.stanford.edu/entries/
olympiodorus/>
Wooten, C. (1987). Hermogenes’ ‘On Types of Style’. Chapel Hill: University
of North Carolina Press.
Worthington, I., and Foley, J. M. (2001). Epea and Grammata: Oral and
Written Communication in Ancient Greece. Leiden: Brill.
Zanker, P. (1996). The Mask of Socrates: The Image of the Intellectual in
Antiquity (A. Shapiro trans.). Berkeley: University of California Press.
Zeller, E. (1883). A history of Eclecticism in Greek philosophy (S.F. Alleyne,
trans.). London: Longmans, Green, and Co.
Zeller, E. (1990). Die Philosophie der Griechen in ihrer geschichtlichen
Entwicklung. Hildesheim: Olms.
Ziai, H. (2005). Recent Trends in Arabic and Persian Philosophy. In
P. Adamson and R. C. Taylor (eds), The Cambridge Companion to Arabic
Philosophy, pp. 405–25. Cambridge: Cambridge University Press.
Zoubos, A. N. (1958). ¯høæ › ºåÆæ . Athena, 62, 194–203.
 General Index
For individual works by ancient authors, see Index of Passages
Academics, Academy 7, 29, 32n.37, 41,
43, 50n.83, 54, 68, 80–7, 95, 205
Old Academics, Old Academy 7, 12,
32, 39–40, 69, 79n.3, 80–2, 86–7,
200, 230
Achaïcus and Sotion 213
Adrastus 7n.17, 14, 38n.53, 201–2,
204, 214
aisthēta see perceptible things
Alcinous 8n.18, 11n.31, 40, 70, 148n.45,
200, 214
Alexander of Aigai 202, 215–6
Alexander of Aphrodisias 4–6, 10n.24,
10n.29, 13, 24, 25n.7, 34n.43,
43n.68, 46n.74, 49n.80, 63,
65n.109, 70, 80, 81n.34, 116n.25,
123, 138, 185n.22, 190–2, 201,
202–7, 215, 216, 227, 231, 237
ambiguity (puzzles/fallacies based
on) 39, 48, 54, 57, 160
Ammonius 6, 14, 24, 25n.7, 28, 34n.43,
39, 60n.98, 66, 164, 166, 213,
217–8, 227
anakinein (actuate, stir up) 9n.22, 57,
64–5, 239
‘Ancient exegetes’ of the
Categories 72, 108
Andronicus of Rhodes 6, 10, 14–6,
21–77, 95, 99, 108, 127, 133, 136,
137, 145n.38, 153, 172, 178,
179n.5, 181–3, 186, 206, 193n.41,
195–9, 201n.5, 206, 211–4, 217,
219–32, 233, 235, 236, 241
on absolute (essential, per se, kath’
hauto) and relative (accidental,
per accidens, pros ti)
predication (see also,
predication) 21–3, 42–57, 68,
74–6, 82–7, 90, 96, 98, 151, 240
on Aristotelian essentialism 10, 17,
42, 46, 85
athetization of De Interpretatione,
31n.32, 38n.55, 188, 227
on Categories as guide for
demonstration (not persuasive
dialectic) 7–12, 21–2, 36–48, 54,
68, 91, 205
on Categories as non-technical/
suitable for beginners 22, 37–8,
42, 57
on logic 10, 31–6, 38–9, 42, 51n.84,
57, 75, 85, 96
Peri Pathōn 29, 232
on privileging Aristotle’s ‘esoteric’
works 25–6, 74
on renaming Aristotle’s Categories 8,
11, 22, 26, 36–8, 201
on reordering Aristotle’s catalogue,
placing Categories at front 7–8,
11, 21, 26, 30–2, 54, 68, 75, 87,
205, 208
‘Roman Edition’ of the Categories 8,
26, 28–9, 219, 223 see also Roman
Edition
Antiochus of Ascalon 11n. 31, 32, 34,
35, 39, 40, 41, 55, 68, 69, 73n.129,
80, 200
Antipater of Tyre 136, 179
Apellicon 8, 26n.12, 29, 220–3
apodeixis, see demonstration
Apollonius Dyscolus 145, 156, 179
aporia (puzzle) 5n. 13, 74, 110–12, 115,
159–62, 180, 181n.11, 185,
186–98
Ps.-Archytas 12, 14, 49n.80, 53, 66,
70, 71n.125, 75, 78–82, 88, 90,
94, 97–9
Arethas 166n.65
Ariston of Alexandria 8, 14, 29, 50,
65, 72–5, 79, 87n.17, 88n.19,
95n.29, 98, 108, 205,
211, 213
Aristoteles (possible teacher of
Alexander of Aphrodisias) 215
Aristotle (for individual works, see Index
of Passages)
274 General Index
Aristotle (for individual works, see Index
of Passages) (cont.)
esoteric works, school treatises 5, 8,
11, 25–6, 29–30, 46n.74, 68, 74,
80, 199, 205, 211, 220–3, 232
Categories
as entitled, Before the Topics 3n.9,
7–8, 11, 22, 36, 62, 76, 201, 214
and metaphysics 1, 55, 75, 88, 91,
95, 99, 127, 148, 159, 162, 170,
192, 201, 205, 211, 217, 237 see
also metaphysics
and ontology 2, 5n.13, 12, 22–3,
43n.68, 51n.84, 53–4, 57n.93,
74–6, 87ff., 95, 99, 103, 105,
107n.9, 113, 127, 130, 143, 178–9,
236–7, 241 see also ontology
and rhetoric 7n.9, 8, 21, 35, 38, 68,
69, 76, 106, 111, 112, 120, 127,
130, 132, 178, 213 see also
rhetoric
and semantics 2, 11n.31, 13, 17, 63,
77, 132–3, 142, 144–5, 172, 188,
198, 206, 212, 235n.4, 239–40 see
also semantics
skopos (target, subject-matter) 13,
36n.49, 55, 66, 88, 93–4, 99, 130,
131n.5, 133, 137, 144n.36, 145,
149, 155, 159, 164–7, 178–9, 187,
196n.45, 202, 206, 213, 215, 216,
218, 235
Arius Didymus 79–80
Artemon 232
Aspasius 12n.33, 14–5, 24, 27, 44n.69,
73, 81n.8, 181, 184, 201, 204, 214,
230, 232
Athenaeus 220, 222
Athenodorus 8, 13, 14, 72n.128, 76–7,
196, 108, 123, 127, 129–39,
143–57, 161–73, 178, 184, 197–9,
204, 205–6, 213, 217, 234
Atticus 106, 111, 114, 115, 127, 200–1, 214
Avicenna 5
Boethius 4, 6, 9n.21, 10n.24, 26 & 28,
21n.2, 15, 23n.3, 25n.7, 29,
35n.44, 37, 38n.53, 40–5, 49,
55–6, 68, 217, 227n.19, 229
Boethus of Sidon 7, 9n.22, 11–3, 17–8,
22, 24n.4, 25n.8, 27–31, 32–3,
40n.60, 51n.84, 59, 62–5, 67–8,
70, 72–77, 95–6, 103–5, 108–17,
118–28, 130–3, 138–9, 143,
145n.38, 149, 172, 177–99,
201n.5, 202, 205–8, 212–5, 226,
230–1, 238–9
on the Relative 7, 9n.22, 18, 64–5, 72,
74, 108, 123–6, 185–6, 190–5, 239
on sensible, perceptible being 12–3,
67, 187–9
on teasing out prolēpseis 64–8, 133
Cassius Dio 139
Categories and categories, see Aristotle,
Categories
Chrysippus 33, 39n.59, 91, 140–3,
145n.38, 148, 156, 162–4, 182, 232,
Cicero 11, 15, 27, 28n.22, 32, 34, 39, 55,
68, 79n.2, 135, 223
concepts, common or innate
conceptions, see ennoiai
Cornutus 8n.19, 77, 96, 123, 127, 129–35,
139–45, 147, 149–57, 161–5,
169–73, 197, 206, 212–13, 217
co-signify (sussemainein), co-signifier
(sussemantikon) 145n.38, 151,
154–7
Crantor 81, 91
Cratippus of Pergamon 27–9, 73n.129, 74
definition 10, 11, 21–2, 39, 40–7, 58, 59,
64–6, 115, 160, 162, 234, 239
Aristotelian 65
and demonstration 55, 76
and division 39–47, 56–7, 69, 84, 206
‘ennoematic’ 65
essential 46
formal 65, 71
scientific 11, 45, 47
technical 64, 178
demonstration 7, 11, 21, 35–43, 47–8,
54–7 68–9, 75–6, 205, 228, 235
see also apodeixis
Dexippus 6, 14, 86n.16, 104, 129, 130,
132, 135–8, 145–6, 149, 154–7,
161–4, 172–3, 191, 203,
215, 217
dialectic 3n.9, 7, 9–11, 21, 33n.39, 35,
36–40, 42, 48, 54–6, 59, 61, 67–8,
72, 77, 82, 91, 99, 132–5, 140–4,
164, 170–1, 205–7, 213, 240
and demonstration 7n.18, 10–11,
21n.1, 35–42, 48, 68, 205
and persuasion 7, 21, 35–6, 54
 General Index
Stoic 39–40, 59, 65, 77, 132–4, 140–4,
164, 171, 206, 213
diaresis (division) 10–2, 21, 35n.44, 37,
39, 40, 41–8, 57, 69, 71, 83–4,
130–5, 143, 145–7, 147–52, 162,
165–72, 187, 205, 229–31, 238
diarthrōsis (articulation, excavation) 9,
59–68, 119, 183, 203, 205
dyad 75, 81, 88–9, 124, 190
Elias 14, 34, 37n.52, 97,166n.65, 218
endoxa 38
ennoiai 56, 59–62, 64, 68, 69n.119, 71,
208 see also concepts, common/
innate conceptions
equivocation (fallacies of) 11, 22, 48,
54–7, 61, 68
essentialism 9–11, 17, 42, 46, 57n.93, 65,
85, 237, 240
Eudemus 24, 231
Eudorus of Alexandria 7, 11–4, 22, 40,
41n.64, 53, 67, 72n.128, 75,
78–97, 99, 127, 200, 205, 211,
213, 217, 228
Eusebius 148n.44, 200, 214
fallacies, see ambiguity
al Farabi 5
Galen 6, 24, 65, 201, 204, 214–6
grammar, grammatical reading of
Categories 2, 8n.19, 47, 76–7,
130–5, 140–5, 156, 164, 170, 178,
213, 233–8
Herminus 9n.22, 12–8, 59, 62–3, 70, 99,
127, 133–4, 143, 149, 167, 180,
183, 187, 195–7, 201–3, 206–7,
212–6, 222, 224n.12
Hermodorus 54, 82n.12
heteronyms 59, 167–8
Hippolytus 24n.6, 70n.123
hupographē (rough sketch, pre-technical
outline, delineation, outline
account) 9, 12, 18, 22, 46–8, 57,
59, 64–6, 93, 178, 183, 239
hupolepsis 24n.6
Iamblichus 6, 14, 25n.8, 36n.49,
70n.123, 86, 88, 93, 104n.4,
105–7, 109n.13, 112, 114–8, 123,
127, 146, 149, 151–2, 168–9,
275
183–4, 195–7, 202, 204, 206–8,
211, 216–8
intelligible (things, realities,
noēta) 12–3, 33, 67, 74–5, 78, 88,
90, 91–4, 96, 106, 107n.9, 110–3,
117–20, 127–8, 155, 178, 189,
190–1, 194, 205–6, 211–3
number 91, 110–1, 123, 124n.39, 127,
189–91
katēgoria see predication
al-Kindi 5
krasis (blend) 24n.6
logic 31–9, 42, 67, 75, 96, 131–3, 139,
146, 167–9, 171, 173, 183–4, 202,
221, 225–8, 231, 235, 237, 239 see
also Andronicus
Lucius (and ‘Lucians’) 12–5, 72, 74–6,
87, 91, 94–6, 99, 103–115,
118–128, 130, 139, 143, 151,
153–6, 163, 168, 172, 178, 180–4,
186, 189–91, 193–8, 205–7,
212–3, 216, 234, 238, 240
Macrobius 141n.24, 177n. 3
metaphysics 1, 55, 75, 88, 91, 95, 99, 127,
148, 159, 162, 170, 192, 201, 205,
211, 217, 237
Moderatus 88, 109n.12&14
monad 75, 81, 88–91, 110, 113, 123–4,
128, 151, 155, 189n.34–191, 193
natural language 2, 17, 76, 141, 148–9
Neleus 220–2
Neoplatonists, Neoplatonism 2n.5, 4, 7,
13–5, 22, 25, 27n.19, 43, 49n.80,
53, 59, 66, 68n.118, 72, 97,
98n.37, 106ff., 130–2, 139, 142,
178n.4, 187, 196, 199, 208, 211–2,
216–18, 220, 226, 228, 235
Nero 139
Nicostratus 12, 14, 39n.59, 59, 74, 87, 91,
94–6, 99, 103–7, 111–8, 120,
124–7, 151, 154, 180–2, 193–8,
206, 213, 216, 238, 240
Old Academy 7, 12, 32, 39–40, 69,
79–82, 86–7, 200, 230
Olympiodorus 6, 14, 39, 166, 217–8
ontology 2, 5n.13, 12, 22–3, 43n.68,
51n.84, 53–4, 57n.93, 74–6, 87ff.,
276 General Index
95, 99, 103, 105, 107n.9, 113, 127,
130, 143, 178–9, 236–7, 241
‘onymies’ 9n.22, 58–61, 95, 144, 146,
158, 165, 168–71, 192, 233–4
homonyms 9n.22, 15–6, 48, 54–9, 85,
96, 111–2, 115, 119, 131, 158–61,
166, 192n.40, 201, 233–4
paronyms 16, 48, 58–9, 112, 115,
164n.63, 166, 192n.40, 233–4, 241
synonyms 16, 54–6, 58–60, 112, 115,
160, 166, 192n.40, 201, 233, 240
opposites, table of 83
organon (tool/instrument) 32n.37, 33,
34n.43
Origen 148
ousia (reality, essence, substance) 2, 10,
13, 21n.2, 22, 42, 45, 48–54, 58,
65, 67, 70, 84–94, 107, 113–5,
117–26, 151–3, 178n.4, 190, 194,
199, 212, 236–40
Panaetius 23, 33
perceptible things, objects, beings 13, 33,
47, 66, 81n.7, 88, 91–93, 128,
178n.4
Peripatetics 10, 14, 18, 32n.37, 34, 35,
37, 44, 46, 68, 73–4, 85, 136,
201–4, 219, 224, 227n.19,
229–30, 237
Persius 139
Philo of Alexandria 39, 53, 55, 82, 88,
90, 99, 207, 212
Philodemus 73n.129, 97, 223
Philolaus 80
Philoponus 9, 25n.7&8, 27, 31–2, 36, 38,
60n.98, 73, 166, 181, 217–8,
225–7
Plato 6, 7n.16, 10n.24, 12, 39–42, 48,
49n.80, 55, 57n.93, 64, 69, 80,
81n.8, 82n.12, 85n.15, 91–3, 97,
122, 131, 145n.38, 148, 200, 201,
207, 213, 230–1
Platonic categories 82–3
Platonists, Platonism 11n.31, 12, 60,
74–6, 80, 82–91, 96, 99, 103–27,
179, 182, 189n.34, 191, 195–9,
200, 205, 212, 214, 217, 238–40
Plotinus 6, 12, 15, 29, 42, 67, 76, 104–7,
110, 112–4, 116–9, 124n.39, 127,
185, 190, 192, 194–7, 198, 206–7,
213, 216, 219, 225n.13, 229, 231
Enneads 30, 94, 104, 107, 110, 116–8,
154, 156, 212, 216, 224–6
Plutarch 8, 15, 24, 26, 28–30, 40–1, 80,
81, 91–3, 108, 109n.12&14, 207,
213, 219–21, 223–4, 226
(the) Point 110, 113, 123–4, 128, 151,
155, 190, 193
polyonyms 167–8
Porphyry 4, 6, 9n.22, 13, 14, 22, 25, 29,
30–1, 35n.44, 36n.39, 38, 42–3,
58–63, 77, 86, 99, 106, 109, 112,
114–23, 126–7, 131–4, 135,
138–40, 143–50, 153–5, 161–9,
173, 180, 183–9, 193–9, 201,
206–7, 215, 216–7, 219–26,
228–9, 231
Isagoge 5, 31, 43n.68, 49n.80, 63, 70,
192n.40, 207, 216, 218, 225
Life of Plotinus 8, 24, 26, 30, 215,
224, 226
On the Categories to Gedalius 14,
25n.7, 109n.13, 115–6, 118n.30,
145–6, 149, 151, 183n.16, 185–6,
188, 195, 197, 207, 216–7
On the Categories by Question and
Answer 11n.31, 24, 37, 49n.80,
63, 69, 74, 99, 109n.13, 146–50,
130, 133, 140, 154, 157, 168,
185–7, 193–4, 195, 197, 203,
213, 225
Posidonius 33, 220, 229n.22
practical philosophy 32n.37, 34, 38
pragmata (realities) 40, 55, 96, 131n.5,
143, 157–8, 187–9, 202–3
predication
absolute (kath’ hauto) and relative
(pros ti) 7, 10, 12, 42, 44n.69,
48–9, 53, 71n.125, 75–6, 82–7,
89, 93–7
accidental 9–10, 22, 42, 44–9, 53, 56,
57n.93, 62, 65, 68, 71, 85, 93, 238
equivocal 16, 56, 59
essential 9, 22, 42, 45–7, 49, 56, 65, 68,
71, 85, 238
homonymous see ‘onymies’
non-substantial 10, 42, 45, 47, 52,
84, 89
per accidens 11, 21, 41, 44, 52, 56,
57n.93, 85, 227n.19, 230
per se 11, 21, 41, 44, 45, 46, 51, 56,
57n.93, 85, 227n.19, 230
 General Index
synonymous 16, 54–6, 58–60, 112,
115, 160, 166, 192n.40, 201,
233, 240
univocal 16, 56, 59
prolēpsis (preconception) 9, 12, 18,
22, 57–68, 75–6, 183,
203–5, 239
Ptolemy the Unknown, catalogue of 30,
220–3, 226, 228, 232
Pythagorean or Neo-Pythagorean
interpretation of Categories 12,
41n.64, 71, 75, 78–99, 108–10,
112, 124n.39, 127, 182n.15,
189n.34, 191, 211–2, 217
quality 2, 17–8, 22, 36n.49, 47, 49–52,
58, 75, 82n.9, 86–90, 94, 99,
107, 109–11, 113, 115, 117,
120–7, 194, 227n.18,
236–9, 241
quantity 2, 9n.22, 17, 22, 49–52, 61–2,
75, 82n.9, 86–90, 99, 106–7, 109,
113, 117, 120–1, 123–7, 178n.4,
191,194, 237–8, 240
Quintilian 8n.18, 38n.56, 76, 137, 205
relation 50–4, 62, 84, 88n.19, 99, 115,
117, 194, 228, 236–41
rhetoric 7n.9, 8, 21, 35, 38, 68, 69, 76,
106, 111, 112, 120, 127, 130, 132,
178, 213
‘Roman Edition’ of the Categories 8, 26,
28–9, 219, 223 see also
Andronicus and Aristotle’s
Categories
277
semantics 2, 11n.31, 13, 17, 63, 77,
132–3, 142, 144–5, 172, 188, 198,
206, 212, 235n.4, 239–40
Sextus Empiricus 32, 139, 142–3,
145n.38, 148n.43, 156
skopos see Aristotle, Categories
Sosigenes 14, 70n.123, 188n.31, 203,
206, 207, 215–6
Speusippus 59, 80, 95, 127, 189n.34,
191–2
Strabo 8, 15, 23–8, 68, 73n.129, 78–9,
108, 135–6, 177, 179–81,
219–24
syllogism 11, 16, 39n.59, 55, 134, 228, 235
Syrianus 98n.37, 168–9, 202, 215
L. Calvinus Taurus 14, 111, 201, 214
Themistius 24, 40, 80, 97, 116n.25
Theophrastus 26–7, 29–30, 219–22
Tyrannio (the grammarian) 8, 28–9, 75,
205, 220, 223–4
use-mention confusion 2
vagueness (correcting) 69, 71, 78
verbal expressions (lexeis katho
lexeis) 12–3, 77, 95, 131, 134
(verbal ambiguities), 143–6,
153n.52, 155, 158, 164, 169, 212,
230n.23
Xenocrates 24n.6, 32, 40, 48–50, 53–4,
80–3, 86, 91, 95, 153, 230–2
Zeno (of Citium) 33, 34n.43, 135–37, 141
 Index of Passages

Alcinous Aristotle
Didaskalikos Categories (Cat.)
6, 159, 43–160, 30: 40 1, 1a1–15: 16–17, 192, 233
Alexander 2, 1a16: 17, 164–5, 234
On the Prior Analytics 2, 1a20–1b9: 17, 42, 47, 238
2, 35–4, 29: 34 3, 1b10–15: 17, 57n.93, 59, 112
160, 28–161, 1: 227 4, 1b5–2a4: 239
402, 1–405, 16: 138 5, 2a12–4b19: 17, 42, 240
On the Topics 6–8, 4b20–11a39: 17–18, 42, 240–1
42, 27–43, 2: 46, 237 7, 6a37: 227n.18
On Meteorology 7, 8a31: 17–18, 52
143, 13: 203 15, 15b17–30: 18, 117, 241
On the Soul On Interpretation (De Int.)
90, 2–15: 190 1, 16a3–8: 142, 188
On the Metaphysics Prior Analytics
242, 15–16: 49n.49, 231 83a22: 162
243, 3: 231 Posterior Analytics (An. Post.)
1.1, 184a17–21: 33
Ammonius 1.4, 73a34-b5: 45
On the Categories (in Cat.) 1.27, 87a26: 191
5, 28: 227 2.13, 96a20–97b2: 45, 84
6, 3–8: 39 Topics
7, 13: 227 1.9: 68
93, 9–12: 60n.98, 66 6.4, 141b26: 44
On the Prior Analytics Sophistical Refutations
(in An. Pr.) 22, 178a4–5: 56, 68
31, 11: 181 Physics
Anonymous 1.1, 184a17–21: 33
On the Theaetetus Metaphysics
col. 46, 43–9: 60 1.6, 988a7–12: 81n.8
On the Categories in the ‘Archimedes 5.5, 1017a23–30: 48n.78
Palimpsest’, see Porphyry 5.6, 1015b16: 230
Apollonius Dyscolus 5.7, 1017a8–30: 53, 230
On Adverbs (De adv.) 5.23, 1023a8: 117
121, 3: 145n.38, 156 7.17, 1041a25–32: 44
On Construction (De const.) On the Good
402, 2: 145n.38 fr. 2 Ross: 190n.34; 191
[Archytas] Arius Didymus
Peri tōn katholou logōn ap. Stobaeum Ecl.
22, 31: 98 2.42, 7: 79
31, 5: 98 Aspasius
31, 3–5: 99 On the Nicomachean Ethics (in Eth.)
31, 6–32, 23: 70–1, 78, 99 44, 24: 27, 73, 181, 232
 Index of Passages 279
Athenaeus 11, 18–20: 161
Deipnosophistae 12, 3–11: 137, 173
1.3a: 220 21, 18–19: 95n.32
5.214d: 220 32, 18–29: 154–6
Aulus Gellius 45, 3–30: 191
Attic Nights (N.A.) Diogenes Laertius (DL)
2.6.1: 141n.24 Lives of the Philosophers
10.4: 148n.43 5.28: 35
20.5: 221 5.22–27: 30, 221, 226
Boethius 5.28: 33–5
On the Categories (in Cat.) 5.53: 220
162A: 97 7.34: 135, 136
263B: 36, 241 7.39–40: 167
On Division (de Div.) 7.40: 132, 169–70
875D-6D: 35; 42–3, 229 7.43: 134
880C-D: 55 7.58: 164
892A: 44, 230 7.62–3: 134, 142
Proem: 229 7.65: 162
7.68: 136–39, 173
Cassius Dio 7.69–70: 137, 161, 162, 163
Roman History 7.72: 161, 163
62.29.2: 139–40 7.134: 162
Charisius 7.192: 143
Art of Grammar (Gramm.) 9.42.4: 136
162, 9: 139 Elias [David]
Cicero On the Categories (in Cat.)
Academica (Acad.) 107, 6: 97
1.19: 32 117, 17–118, 31: 34
1.30–32: 34 160, 20: 204n.18
2.91–92: 34 241, 30: 37
1.3–2: 39 Epicurus
1.5: 39 Letter to Herodotus (Ep. Hdt.)
On Duties (De off.) 75: 149n.45
1.1: 27–8
1.1.1: 27–8 Eusebius
2.24: 179n.7 Praeparatio Evangelica (PE)
3.25: 27–8 6.8.1–10: 148n.44
1.2: 28 11: 200–1
On the Orator (De orat.) 16; 200–1
2.36.152: 223 Galen
Cornutus Institutes of Logic (Inst. log.)
Summary (Epidrome) 13: 204
2.4, 2.9, 65.8: 141n.27 On My Own Books (De libr.
propr.)
Dexippus 19, 42–3: 204, 204
On the Categories (in Cat.) 42, 20–43, 1: 201, 2014
1, 9: 130, 157
5, 16–22: 129 Lucian
7, 1: 188n.31 Life of Demonax
7, 4: 203, 215 56, 1–6: 203n.11
11, 1–4: 157 [Lucian]
11, 4–16, 13: 156–61 Long Lives (Macrob.)
11, 7–12: 154–6 21: 136
280 Index of Passages
Lucretius 6.1.5: 154, 156
On the Nature of Things 6.1.23: 117, 118
4.824: 149 Plutarch
Macrobius On the Creation of the Soul in the
On the Dream of Scipio Timaeus (de proc. an.)
1.14.19: 177n.3 1013B, 1019E, 1020C: 80
Olympiodorus 1023D-E: 41, 91–3
On the Categories (in Cat.) Table Talk (Quaest. Conv., QC)
9, 9–11: 59 7: 109n.14
8.7–8: 108
Origen Life of Cicero
Against Celsus 20.3: 109
1.24: 142, 148 Life of Cato the Younger
Exhortation to Martyrdom (Exh. 10, 1–3: 136
Mart.) 16, 1: 136
46: 148n.43 Life of Sulla
Persius 26.1–2: 8, 24, 26, 28, 30, 220,
Satires 224, 226
5: 139 Porphyry
Philodemus Life of Plotinus
Index Acad. Herc. 24: 26, 30–1, 219, 224–5
col. 6: 97 Introduction (Isagoge)
col. 35: 73n.129 1, 2–9: 31
Philoponus 1, 3–17: 49n.80, 225
On the Categories (in Cat.) 1, 18–2, 10: 192n.40
5, 15–20: 27, 31–32, 36, 38, 73, 181, 6, 10–16: 70
225, 226 On the Categories By Question and
27, 18–26: 227 Answer (in Cat.)
45, 8–12: 227 55, 3: 63
167, 21–168, 3: 60n.98 56, 6–13: 144
56, 22–31: 37
Plato 58, 3–59, 14: 162
Cratylus 58, 3–7: 147–8, 187
390A: 148n.45 58, 5–13: 69, 144
423E: 148n.42 58, 9–20: 99
424B: 148n.42 58, 30–59, 2: 143, 147–8
Parmenides 59, 3–14: 142–44, 146, 149, 150
128E-189B: 85n.15 59, 16–33: 59, 63, 143, 183,
Phaedo 203, 225
100C: 85n. 15 59, 34: 167
Phaedrus 59, 17–33: 31, 63, 133–34,
265E: 131 149, 203
Sophist 60, 1: 168
218C: 55 73, 15–28: 118–9
255C12-D7: 48, 230–1 86, 20–22: 130, 132, 135, 140,
Symposium 155, 163
211B: 85n.15 86, 33–4: 197
Timaeus 103, 18: 194
37A-C: 91–2 111, 9–112, 5: 194, 213
Plotinus 125, 14–23: 50n.82, 52
Enneads 125, 30: 74
6.1: 104, 107 154, 7–12: 61
6.1.2: 117 154, 12–3: 62
 Index of Passages 281
On the Categories To Gedalius [?] (ed. 16, 1–4: 214
Chiaradonna, Rashed, Sedley & 16, 14: 202
Tchernetska 2013) 17, 3–7: 150
1, 26–3, 1: 39n.59, 59 18, 22–19, 9: 131, 133, 142, 144, 146,
2, 26–31: 106 150, 162, 163, 165, 166–71,
6, 9–7, 7: 57n.93 172–3
14, 4: 192n.40 19, 1: 131
Proclus 19, 7–9: 168
On the Cratylus (in Crat.) 19, 9: 146
16.18: 148n.45 19, 27: 168
18.1–4: 148.n45 21, 2–7: 115
On the Parmenides (in Parm.) 21, 2–25, 9: 96, 112, 115
1083, 37: 200n.2 21, 15–22, 1: 58, 61
21, 2–7: 115
Quintilian
21, 22–4: 95n.32
Institutes of Oratory (Inst. Orat.)
22, 1–2: 168
2.17.15: 137
23, 13–15: 168
3.23–34: 76, 205
23, 25–34, 5: 168
Sextus Empiricus 23, 29: 111
Outlines of Pyrrhonism (P.H.) 24, 6–25, 9: 55–6
2.104: 142 26, 17–30, 3: 32, 62n.103, 95n.32
2.176, 5: 145n.38, 156 26, 22: 111
3.267–8: 148n.43 29, 5: 95
Against the Mathematicians (Adv. 29, 23–30, 5: 96, 115, 125n.43, 183
Math.) 30, 3–5: 95n.32
1.143–4: 148n.43 30, 16–22: 115, 127, 201
7.16: 32 32, 19: 201
8.11–12: 142 38, 19–23: 95
8.223: 161 40, 6: 148n.43
11.241–2: 148n.43 41, 14–9: 188–9
Simplicius 41, 28: 187
On the Categories (in Cat.) 48, 1: 106, 180
1, 3–2, 29: 231 48, 1–50, 9: 113, 115, 118, 120–22, 193
1, 8: 177 50, 2: 180
1, 18–2, 2: 103, 183 54, 8–21: 230, 231, 236
1, 19: 106, 112, 180 58, 14–22: 104, 118, 124–25, 127,
1, 22: 107 181n.11
1, 8: 177 58, 23–8: 125, 188n.30, 193n.42
1, 19–20: 96 58, 27–59, 4: 186, 193
2, 3: 107 59, 5–14: 131, 154
2, 4: 185 59, 17–18: 196
2, 9: 107 62, 7–23: 99
3, 2–10: 116, 208 62, 24–63, 3: 163
6, 4–5: 38 62, 24–7: 76, 131, 132, 144, 146, 151,
7, 4–9, 24: 215, 302 152, 161, 170, 172
7, 23–31: 104, 183 62, 28–65, 14: 106, 118, 127
10, 19–20: 215 62, 30–1: 151
11, 23–9: 177, 183, 187, 198 63, 2: 185, 197
12, 19–13, 12: 68n.118 63, 4–67, 9: 151–2, 162
13, 11–18: 202, 168, 187, 215 63, 4–64, 12: 151
13, 22: 207 63, 21: 231
15, 26–16, 5: 37, 201, 202 63, 22–6: 45, 48, 81, 230
282 Index of Passages
Simplicius (cont.) 201, 34: 50n.82
63, 28–30: 49 202, 1–5: 231
63, 33–64, 1: 85–6 202, 2: 98
64, 13–65, 14: 123, 125, 153, 154, 193 202, 5: 62, 65n.110, 227n.18, 228, 231
64, 13–66, 15: 151 203, 1: 213
64, 18–65, 13: 106, 155, 172 203, 4–6: 228, 231
65, 13–66, 21: 162 206, 10–15: 87, 228
65, 13–24: 110, 113, 123–4, 190, 231, 19–23: 104, 116
113, 123–4, 189, 190, 194 263, 19–22: 227n.18
66, 16–31: 152 268, 19–32: 104, 116
66, 32–67, 9: 152, 172 270, 2: 231
73, 15–76, 16: 91, 96, 110, 111, 113, 302, 16: 192
115, 116, 118–20, 127 339, 19: 183n.16
73, 28: 106, 107, 194 342, 21–5: 49, 183n.16, 227n.18
74, 3–17: 66, 178 347, 19–348, 2: 194, 231
75, 15–28: 118–9 348, 2–7: 192–3, 194
75, 28–31: 65–6 357, 28: 227
76, 13–18: 117, 119 359, 4: 161
76, 20: 98 368, 1–371, 27: 111–13, 116–17, 127
78, 4–20: 119–20, 178, 183n.16, 370, 1: 104, 106
189–90, 191, 196 370, 5: 117
82, 2–20: 67, 178 370, 8: 104
125, 16: 106, 124, 194 373, 8: 192
127, 11–128, 10: 107, 116, 127 379, 9–20: 36–7, 60, 228
127, 30: 106 381, 3–32: 106, 116
128, 7: 131 390, 15: 106
129, 1: 131 410, 24: 104
129, 27: 110 414, 23–415, 15: 116
134, 5: 227n.18 428, 3: 104
149, 14: 196n.46 On the Physics
154, 3–12: 61 99, 24–31: 231
154, 12–13: 62 122, 33: 201n.6, 214
156, 16: 118, 122n.36 924, 18: 223
156, 16–25: 106, 109, 113, 126, 194 1036, 18: 223
157, 23–158, 27: 195 On On The Heavens (in de Cael.)
159, 10–15: 64, 183, 195 153, 16: 215n.8
159, 14–202, 1: 185, 192 366, 25: 185n.22
159, 25: 203 430, 32: 215
159, 32–3: 72, 108, 231 674, 4: 185n.22
163, 3: 196n.46 Stephanus
163, 6: 185, 192 On On Interpretation (in De Int.)
163, 15: 196n.46 9, 19–22: 148n.45
163, 19–20: 183n.16 10, 7: 148n.45
163, 28–9: 65
166, 17–19: 50 Stobaeus
174, 14–26: 84, 89–90 Extracts (Ecl.)
181, 7–30: 81, 84 2.42, 7: 79
174, 14–175, 11: 53 Strabo
181, 10: 79 Geography
181, 22–30: 89 2.3.5: 28
187, 7: 148n.43 13.1.54: 220
187, 24–34: 133, 170 14.2.13: 23
 Index of Passages 283
14.5.14: 135, 136, 179n.5 Themistius
16.2.24: 27, 179 On Aristotle On the Soul (in An.)
17.1.5: 79 31, 1–32, 34: 80
Syrianus Xenocrates
On the Metaphysics Fragmenta (ed. Heinze)
100, 6: 215 12: 7n.16, 82